Cross-Reference to Related Applications

[0001] The present application claims priority from Australian Provisional Patent Application No 2022901848 filed on 30 June 2022, the contents of which are incorporated herein by reference in their entirety.

Technical Field

[0002] This disclosure relates to detecting and quantifying a protein in a sample.

Background

[0003] Within the increasingly globalised agri-food supply chain, consumers have access to limited information on the origin of their food due to the complexity of the food production chain. The detection, quantification, and mitigation of contamination is essential to ensure food safety. Allergens pose a significant health burden to the global population, with 3.5-4% being affected by food allergies. Those with allergies mainly adhere to diets to avoid allergic reactions. This means food products are to be properly tested and labelled for the presence of known allergens.

[0004] In the European Union, the United States and Australia, only foods that have been intentionally incorporated into food products must be listed as ingredients. As such, trace amounts of allergens that have unintentionally entered the product (e.g., through cross-contamination) do not need to be declared. However, to protect consumers from allergic reactions and themselves from legal consequences, many food manufacturers voluntarily employ precautionary allergen labelling (PAE) systems. Unfortunately, as PAL systems are unregulated and largely overused, many consumers choose to ignore these warnings, which can lead to dire consequences including lifethreatening anaphylaxis. [0005] Any discussion of documents, acts, materials, devices, articles or the like which has been included in the present specification is not to be taken as an admission that any or all of these matters form part of the prior art base or were common general knowledge in the field relevant to the present disclosure as it existed before the priority date of each of the appended claims.

[0006] Throughout this specification the word “comprise”, or variations such as “comprises” or “comprising”, will be understood to imply the inclusion of a stated element, integer or step, or group of elements, integers or steps, but not the exclusion of any other element, integer or step, or group of elements, integers or steps.

Summary

[0007] A method for detecting a protein in a sample, the method comprising: receiving a sampling device containing the sample, the sampling device comprising an absorbent material with a test area configured to provide a visual indication of the protein in the sample; capturing image data of the sampling device containing the sample, the image data capturing the visual indication; extracting the sample from the sampling device; and detecting the protein using liquid chromatography mass spectrometry (LC- MS) analysis of the extracted sample to confirm detection of the protein as visually indicated by the image data.

[0008] It is an advantage that the sampling device can be used for both the collection and extraction of a sample containing a protein, as the sampling devices can be produced at relatively low costs and can be easily used. The sampling devices can also be used to collect and transport the sample from first location to a second location that is remote from the first location, for LC-MS analysis.

[0009] In some embodiments, extracting the sample from the sampling device comprises chemically extracting the sample to produce an extracted sample using an extraction solution and a first spin filter to produce a supernatant containing the extracted sample.

[0010] In some embodiments, chemically extracting the sample from the sampling device comprises transferring the supernatant to a second spin filter and applying a digestion solution to obtain the extracted sample from the supernatant.

[0011] In some embodiments, the method further comprises detecting a false negative or a false positive by the sampling device using the LC-MS analysis to detect the protein below a limit of detection of the sampling device.

[0012] In some embodiments, the sampling device is provided with a machine readable unique identifier and the machine readable unique identifier comprises one or more of: a barcode; a QR code; and a serial number.

[0013] In some embodiments, the protein is part of: an allergen; an antibody; an antigen; a toxin; or a pathogen.

[0014] In some embodiments, the sampling device comprises a lateral flow device.

[0015] In some embodiments, the sampling device comprises a sampling pad configured to receive the sample and the method comprises extracting the sample from the sampling pad for the LC-MS analysis. [0016] In some embodiments, the sample is collected from a first location using the sampling device and the LC-MS analysis is performed at a second location remote from the first location.

[0017] In some embodiments, the method further comprises: calculating an intensity ratio from the image data by comparing a first intensity of the test area of the absorbent material with a second intensity of an area of the absorbent material other than the test area; and calculating a quantity of the protein present in the sample based on the intensity ratio.

[0018] In some embodiments, capturing the image data comprises capturing the image data with a camera integrally contained in a communication device.

[0019] In some embodiments, the method further comprises creating calibration curves using multiple calibration samples, wherein the calibration samples contain known concentrations of the protein.

[0020] In some embodiments, one of the calibration curves is stored in the communication device for use with sampling devices of multiple pre-defined types.

[0021] In some embodiments, the calibration curves are created and stored on a server device.

[0022] In some embodiments, the method further comprises correcting the image data using a correction ratio calculated based on an average of the first intensity and an average of the second intensity.

[0023] In some embodiments, correcting the image data comprises detecting the visual indication of the protein in the sample in the image data to thereby detect the protein in the sample from the image data. [0024] In some embodiments, the image data further comprises data indicative of one or more of: a GPS location; a time stamp; and the machine readable unique identifier uniquely associated with the sampling device.

[0025] In some embodiments, the image data is in a RAW image format.

[0026] In some embodiments, the method further comprises determining a quantity of the protein in the sample using the LC-MS analysis.

[0027] In some embodiments, the sampling device provides an internal standard, and the method comprises using the internal standard in the LC-MS analysis to determine the quantity of the protein in the sample using the LC-MS analysis.

[0028] In some embodiments, the internal standard is a heavy carbon isotope standard.

[0029] In some embodiments, the method further comprises updating the calibration curves at the intensity ratio in the image data with the quantity of the sample determined using the LC-MS analysis.

[0030] In some embodiments, the method further comprises determining peptide sequence information of the protein using the LC-MS analysis.

[0031] In some embodiments, the method further comprises detecting the presence of a second protein using the LC-MS analysis to verify a purported origin of the sample.

[0032] A system for detecting a protein in a sample, the system comprising: a sampling device configured to receive the sample, the sampling device comprising an absorbent material with a test area configured to provide a visual indication of the protein in the sample; and a liquid chromatography mass spectrometry (LC-MS) system configured to perform LC-MS analysis to detect the protein in the sample extracted from the sampling device, wherein the LC-MS analysis confirms detection of the protein as visually indicated by the image data.

[0033] A method for quantifying a protein in a sample, the method comprising: capturing image data of a sampling device containing the sample with a camera of a communication device, wherein the sampling device comprises an absorbent material and is configured to provide a visual indication of the protein in the sample; the sampling device is provided with a machine readable unique identifier that is uniquely associated with the sampling device; the sampling device provides an internal standard; and the image data comprises data indicative of the machine readable unique identifier to identify the sampling device from the image data, and the visual indication of the protein in the sample; calculating, by the communication device, a quantity of the protein in the sample based on the visual indication in the image data; sending the image data to a server device; performing, by a server device, the steps of: receiving the image data; confirming the identity of the sampling device using the data indicative of the machine readable unique identifier in the image data extracting the sample from the sampling device chemically using one or more spin filters; and in response to confirming the identity of the sampling device, confirming the quantity of the protein in the sample based on the visual indication in the image data by using the internal standard in liquid chromatography mass spectrometry (LC-MS) analysis of the sample extracted from the sampling device.

[0034] It is an advantage to calculate the quantity of the protein in the sample based on the visual indication in the image data to provide a preliminary quantification result, then confirming this preliminary result by using the internal standard in liquid chromatography mass spectrometry (LC-MS) analysis of the sample. The preliminary result is useful in situations where a quick result or approximation is needed, which can later be confirmed by the LC-MS analysis.

[0035] A system for quantifying a protein in a sample, the system comprising: a sampling device, wherein the sampling device comprises an absorbent material and is configured to provide a visual indication of the protein in the sample, the sampling device is provided with a machine readable unique identifier that is uniquely associated with the sampling device, and the sampling device provides an internal standard; a communication device comprising a processor, wherein the processor is configured to: capture image data of a sampling device containing the sample with a camera of the communication device, wherein the image data comprises data indicative of the machine readable unique identifier to identify the sampling device from the image data, and the visual indication of the protein in the sample, calculate a quantity of the protein in the sample based on the visual indication in the image data, and send the image data to a server device; the server device configured to: receive the image data from the communication device, and confirm the identity of the sampling device using the data indicative of the machine readable unique identifier in the image data; and a liquid chromatography mass spectrometry (LC-MS) system configured to: perform LC-MS analysis on the sample, and in response to the server device confirming the identity of the sampling device, confirm the quantity of the protein in the sample based on the visual indication in the image data by using the internal standard in LC-MS analysis of the sample chemically extracted from the sampling device. [0036] A method for detecting a protein in a sample, the method comprising: capturing image data of a sampling device containing the sample with a camera of a communication device, wherein the sampling device comprises an absorbent material and is configured to provide a visual indication of the protein in the sample; the sampling device is provided with a machine readable unique identifier that is uniquely associated with the sampling device; the sampling device provides an internal standard; and the image data comprises data indicative of the machine readable unique identifier to identify the sampling device from the image data, and the visual indication of the protein in the sample; processing, by the communication device, the image data to detect the visual indication in the image data to thereby detect the protein in the sample using the visual indication in the image data; sending the image data to a server device; performing, by a server device, the steps of: receiving the image data; confirming the identity of the sampling device using the data indicative of the machine readable unique identifier in the image data; extracting the sample from the sampling device chemically using one or more spin filters; and in response to confirming the identity of the sampling device, detecting the protein in the sample using the internal standard in liquid chromatography mass spectrometry (LC-MS) analysis of the sample extracted from the sampling device to confirm detection of the protein as visually indicated in the image data and determining a quantity of the protein using the LC-MS analysis.

[0037] It is an advantage to process the image data to detect the visual indication in the image data to thereby detect the protein in the sample using the visual indication in the image data as the visual indication of the sampling device may not always be visible by the naked-eye. The visual indication may also provide a preliminary detection of the protein in the sample. The preliminary result is useful in situations where a quick result is needed, which can later be confirmed and quantified by the LC- MS analysis.

[0038] A system for detecting a protein in a sample, the system comprising: a sampling device, wherein the sampling device comprises an absorbent material and is configured to provide a visual indication of the protein in the sample, the sampling device is provided with a machine readable unique identifier that is uniquely associated with the sampling device, and the sampling device provides an internal standard; a communication device comprising a processor, wherein the processor is configured to: capture image data of a sampling device containing the sample with a camera of the communication device, wherein the image data comprises data indicative of the machine readable unique identifier to identify the sampling device from the image data, and the visual indication of the protein in the sample, process the image data to detect the visual indication in the image data to thereby detect the protein in the sample using the visual indication in the image data, and send the image data to a server device; the server device configured to: receive the image data from the communication device, and confirm the identity of the sampling device using the data indicative of the machine readable unique identifier in the image data; and a liquid chromatography mass spectrometry (LC-MS) system configured to: perform LC-MS analysis on the sample, and in response to the server device confirming the identity of the sampling device, detect the protein in the sample by using the internal standard in LC-MS analysis of the sample chemically extracted from the sampling device to confirm detection of the protein as visually indicated in the image data and determine a quantity of the protein using the LC-MS analysis. [0039] Software that, when installed on a communication device and executed by the communication device, causes the communication device to perform the above method or part thereof.

[0040] Software that, when installed on a server device and executed by the server device, causes the server device to perform the above method or part thereof.

Brief Description of Drawings

[0041] Fig. 1 illustrates a method for detecting a protein in a sample.

[0042] Fig. 2a illustrates an absorbent material of a lateral flow assay before application of a sample.

[0043] Fig. 2b illustrates the absorbent material after the sample has been applied to the sampling pad.

[0044] Fig. 2c illustrates the absorbent material after the lateral flow process.

[0045] Fig. 3a illustrates a method for determining a quantity of a protein in a sample.

[0046] Fig. 3b illustrates a method for detecting a protein in a sample.

[0047] Fig. 4 illustrates an example system for performing one or more of the methods described herein.

[0048] Fig. 5 illustrates different calibration curves for relative gold nanoparticle (GNP) concentration created using image data from a communication device.

[0049] Fig. 6 illustrates calibration curves of GNP concentration and intensity in colour channels of JPEG and RAW images captured using Android and iPhone mobile phone models at 1/60 shutter time. [0050] Fig. 7 illustrates smartphone based quantification of commercial lateral flow assays.

[0051] Fig. 8 illustrates a smartphone application for colour quantification.

[0052] Fig. 9 illustrates optimisation of the LC-MS extraction protocol.

[0053] Fig. 10 illustrates the direct quantification of allergenic peptides from LFAs by LC-MS.

Description of Embodiments

[0054] Targeted proteomic approaches show great promise for laboratory-based quantification of allergens, because such methods do not suffer from cross-reactivity issues, have excellent multiplexing potential, and, can be used as confirmatory analysis to screening tests, which further improves food safety.

[0055] Such confirmatory analysis include the use of liquid chromatography mass spectrometry (LC-MS), which combines the physical separation capabilities of liquid chromatography (or high-performance liquid chromatography) with the mass analysis capabilities of mass spectrometry (MS). While liquid chromatography (LC) separates mixtures with multiple components, mass spectrometry provides spectrometric information that may help to identify (or confirm the suspected identity of) each separated component. In targeted, bottom-up LC-MS-based proteomics, complex mixtures of proteins are first subjected to enzymatic cleavage to produce peptides. The resulting peptides travel down the LC column, eluding based on their physicochemical properties, and are subsequently analysed using a mass spectrometer. Mass spectrometry also provides information about the mass of the target and the mass of its fragments and may also provide peptide sequence information leading to the identification of proteotypic peptides (peptides that are unique for a given protein). However, LC-MS approaches tend to be time consuming and are performed in a laboratory, which may be located remotely from a food production facility, for example.

[0056] Lateral flow assays (LFA) are valuable on-site tests. These assays eliminate the need for specialist equipment and expertise required for traditional laboratory methods. LFAs can enable food manufacturers to test individual batches for crosscontamination. LFAs can empower consumers to monitor the safety of their own food as well and democratise analytical chemistry protocol execution. LFAs are primarily qualitative, reporting the presence or absence of a target analyte. A crucial limitation of LFA interpretation is the reliance on visual inspection, such as naked-eye readings. Variable environmental conditions, such as low levels of ambient light and glare, on top of perceptual differences between users make these assays highly susceptible to misinterpretation. This effect is compounded if a user has impaired colour perception or is longsighted. For these reasons digital readers for the interpretation of LFAs have been developed. For example, the Raptor system (Neogen) enables the quantitative interpretation of a range of test strips (e.g., LFAs for peanut, milk, sesame, almond, gliadin and egg and soy determination). However, such systems only work for LFAs of the specific manufacturer and are costly. It would be beneficial to create a cost-efficient quantification method that is compatible with many commercial LFAs to improve cross-platform compatibility and accessibility of these tests.

[0057] LC-MS are developed for the detection of allergens from raw material, therefore it is difficult to confirm LFA results directly by LC-MS since LC-MS would use a new extraction from novel raw material. Developing an LC-MS approach that enables direct confirmation of an LFA result would thus be beneficial for LFA assay developers (to confirm assay selectivity) and LFA users and regulatory bodies alike (to assure that the assay was performed correctly and confirm LFA test results in a disputed case result, such as a true positive, a false negative or a false positive result).

[0058] Smartphone-based quantification of the LFA results is also appealing because smartphones remove costs associated with additional hardware, such as the digital readers that are specific for a particular LFA. Moreover, smartphones have the capacity to enable real-time geotagging of test locations, secure data storage and management via the cloud and rapid communication to stakeholders. However, smartphone-based LFA quantification has proven challenging with model-dependent spectral sensitivity and large inter-phone variations observed.

[0059] This disclosure proposes a new pipeline and methods for compliance testing using smartphone-based on-site screening and rapid confirmatory analysis. In particular, this disclosure is directed towards the detection and quantification of a protein in a sample from a sampling device, such as a LFA. The presence of a particular protein (or the constitutes of a protein, as known as ‘peptides’) in a sample can be used as a detection method of an allergen or other protein in the sample, such as the presence of the allergen in food. While the disclosed methods are directed towards the detection and quantification of a protein to determine the presence of an allergen in a sample, the disclosed methods are applicable for different types of proteins. For example, the protein may be part of an allergen, an antibody, an antigen, a toxin or a pathogen. Therefore the methods in this disclosure have many applications, and are not simply limited to the food industry.

[0060] The methods in this disclosure relate, particular, to two aspects:

• Smartphone-based quantification of colorimetric test strips such as those found in LFAs, and

• LC-MS based verification of the quantitative or qualitative LFA results by directly extracting and quantifying target proteins from the LFA.

[0061] In an embodiment, a preliminary detection of the protein in a sample is performed by collecting the sample using a sampling device, such as a LFA, at an onsite location. The sampling device gives a visual indication of the presence of the protein in the sample. The preliminary detection of the protein in the sample may also be additionally confirmed by capturing an image of the sampling device using a communication device, such as a smartphone. The communication device may be configured to determine the presence of the protein in the sample using the image captured by the device to analyse the visual test line of the sampling device. Additionally, the communication device may be configured to determine a preliminary quantity of the protein in the sample using the image captured by the device. The sampling device containing the sample can then be sent to another facility or location for analysis of the sample contained in the sampling device using LC-MS. As such, the sampling device serves as a means of transporting the sample, as the sample can be directly extracted from the sampling device. LC-MS of the sample determines the presence and quantity of the protein in the sample, which can be used to confirm the preliminary results of the sampling device at the on-site location.

[0062] Smartphone-based detection is achieved with comparable sensitivity as when performed with the naked eye for all target assays. Quantification was achieved with varying degrees of prediction error (ranging between 15% and 50% depending on the assay). This is a considerable improvement to the art where only a present or absent determination can be made.

[0063] The LC-MS methods show very promising results with detection limits well below the described detection limits, or limit of detection, of the screening tests. Moreover, additional peptides can be detected in the LC-MS method to verify the test was conducted in the correct sample environment (or matrix) and, the LC-MS method can be used to verify if a test was conducted correctly and may be used as a valuable control method.

[0064] Overall, the disclosed methods may facilitate acquiring export compliance documentation regarding allergens, pathogens or protein toxins whereby the smartphone -based screening can be used to generate swift compliance documentation and the LC-MS check can be used to enable occasional controls to check that LFA tests were conducted correctly (by detecting typical peptides found in the matrix as well as the target compound). This can be of value to food exporters/producers and clearly aligns with the Trusted Agrifood Exports (TAE) missions’ goal to increase market access by developing new approaches to meeting the regulatory requirements of trading partners. [0065] As smartphones and other image sensors are evolving, it is possible to retain raw (e.g., DNG; .arw; .crw; RW2; .RAF; .CR2) images on smartphones which has the potential to improve the performance of LFA quantification with such digital devices. Moreover, ratio-metric corrections for ambient light variation has the potential to be more efficient for this approach due to the limited processing performed on this image format.

[0066] As such, this disclosure uses DNG smartphone images and ratio metric ambient light correction for colour quantification of LFAs via a smartphone followed by subsequent LC-MS analysis directly on extracts obtained from the LFAs. Gold nanoparticles (GNPs), a common LFA label, were used to establish the optimal camera settings for assay photography. The performance of RGB colour channels and grey scale from JPEG images were compared with raw intensity values from DNG images under various lighting conditions. The best system was used for the quantification of commercial allergen LFA test strips for gluten (Gluten Protein Rapid Kit, 3M) and peanut (AgraStrip®, Romer; Peanut Protein Rapid Kit, 3M) allergen and the method performance was thoroughly validated. Additionally, a targeted LC-MS method enabling the detection of allergen specific peptides directly from the LFAs was developed and validated. The LC-MS approach enables the direct validation of the LFA results by quantifying allergen specific peptides directly from the LFA. Opposed to completing a separate LC-MS assay, this novel approach enables the validation of any quantitative readings made from a mobile device and also mitigates against sample management issues.

[0067] These methods enable the enforcement of quantitative labelling for trace-level food allergen contamination of food products. Implementing such a system would greatly benefit food safety and protect people with food allergies from unintended allergen ingestion. Moreover, the system has excellent potential to be implemented for the quantification/control of other lateral flow assays in a variety of fields. Overall, the disclosed methods provide an evidence -based protocol that can be used for quicker and more cost-efficient compliance certification for the absence of protein-based targets such as allergens, pathogens, and protein-toxins (e.g., Shiga toxin). This directly aligns with the goal to improve market access to high-value markets as the system can give importing countries stronger assurances around food safety standards, that otherwise form a barrier to trade.

[0068] Fig. 1 illustrates a method 100 for detecting a protein in a sample. In an example, method 100 may be performed by a single person, such as a food safety technician or a laboratory technician, or may be performed by multiple people. Firstly, method 100 comprises collecting 101 the sample using a sampling device. The sampling device may comprise a lateral flow device, such as a lateral flow assay (LFA), lateral flow immunoassay (LFI), rapid antigen test (RAT), or a dipstick. In other examples, the sampling device may be any device configured to collect the sample for LC-MS analysis. The sampling device may also be any commercially available lateral flow device, or may be a lateral flow device that is developed in-house or by a research facility. In an example, the sampling device may comprise an absorbent material, that absorbs the sample and thereby, collects 101 the sample for detection of the protein. The absorbent material may be a paper material, made of cellulose, as an example, noting that other absorbent materials, including synthetic materials, are equally usable. The sampling device may further comprise a housing that holds the absorbent material.

[0069] Fig. 2a illustrates an absorbent material 200 of a lateral flow assay before application of a sample 201. The sample 201 contains the protein 202, which is to be detected using the absorbent material 200 by performing method 100. Collecting the sample 101 occurs by bringing the sample 201 in contact with the absorbent material 200 at the sampling pad 203. As such, the sampling device may comprise a sampling pad 203 configured to receive the sample 201. The absorbent material 200 further comprises a conjugate pad 204, which contains multiple labelled conjugates 205. Each labelled conjugate 205 comprises a label (represented by the small black circles), such as a gold nanoparticle, for example, as well as a binding body (represented by the Y- shaped object). The label of the labelled conjugate 205 produces a colour that can be visually seen by a human, which is used to provide a visual indication of the detection of the protein in the sample. However, in some examples, the label may be a fluorescent or magnetic labelled particle, that would use an electronic reader to assess the result. The binding body is a conjugate molecule to the protein 202 being detected, in the sense that, the binding body may bind to the protein 202. In an example, if the protein 202 is part of an antigen, the binding body would the conjugate antibody.

[0070] The absorbent material 200 further comprises a test area 206 and a control area

207. The test area 206 contains the same or similar binding bodies 208 to the labelled conjugates 205 and therefore, are able to bind to the protein 202. As the labelled conjugates 205 attach, with the protein 202, to the binding bodies 208 of the test area

208, the labels of the labelled conjugates 205 provide a visual indication of the presence of the protein 202 in the sample 201. The control area 207 contains binding bodies 209 that bind to the binding body of the labelled conjugates 205, rather than the protein 202. As the labelled conjugates 205 attach to the binding bodies 209 of the control area 207, the labels of the labelled conjugates 205 provide a visual indication that the sampling device is working correctly.

[0071] The absorbent material 200 also further comprises a wicking pad 210, which is an absorbent pad designed to draw the sample 201 from the sampling pad 203 and initiate flow of the sample 201 through the absorbent material 200. The flow of the sample 201 through the absorbent material 200 terminates at the wicking pad 210, where the wicking pad 210 also acts as a waste collection site. The absorbent material 200 further comprises a blank area 211, which only comprises the absorbent material. The blank area 211 comprises multiple areas: one area defined between the conjugate pad 204 and the test area 206, one area defined between the test area 206 and the control 207, and one area defined between the control area 207 and the wicking pad 210. The blank area 211 is generally the region of the absorbent material 200 where no binding occurs and is generally white to a user who visually inspects the absorbent material 200. The sampling pad 203, conjugate pad 204 and the wicking pad 210 may each be a separate or individual pad that attaches to the absorbent material 200. The absorbent material 200 may be fixed to an inert backing material.

[0072] The sampling device may comprise housing configured to hold the absorbent material 200. The housing may be configured with a window, such that a user may visually see the test area 206 and the control area 207 through the window, without needing to remove the absorbent material 200 from the housing. The window may also show all or part of the blank area 211. The housing may also comprise an opening over the sampling pad 203, that may be used to receive the sample 201 using the sampling pad 203 without removing the absorbent material 200 from the housing.

[0073] It should be noted that in Figs. 2a-2c, only three protein molecules, labelled conjugates, and binding bodies in the test and control area are shown for illustration purposes only and any number of these molecule may be present in reality. However, in general, there are more labelled conjugates 205 than protein 202 molecules. The excess of unbound labelled conjugates 205 are used to bind to the binding bodies 209 of the control area 207, thereby providing a visual indication to the user that the sampling device is working correctly. It is noted that the presence of a visual indication in the test area 206 and the absence of a visual indication in the control area 207, indicates that the sampling has not worked correctly. In an example, the labelled conjugate 205 may only bind to the binding bodies 208 of the test area 206 and not to the protein 202 or the binding bodies 209 of the control area 207. Therefore, this would indicative that lateral flow of the sampling device was not performed correctly.

[0074] Fig. 2b illustrates the absorbent material 200 after the sample 201 has been applied to the sampling pad 203. After the sample 201 has been applied to the sampling pad 203, the sample 201 flows through the absorbent material 200, as indicated by the flow direction 221. The flow of the sample 201 through the absorbent material 200 may also be referred to as lateral flow. As such, Fig. 2b illustrates the absorbent material 200 during the lateral flow of the sample 201 through the absorbent material 200. As the sample 201 flows through the absorbent material 200, the protein 202 molecules in the sample 201 move from the sampling pad 203 through the conjugate pad 204, which allow the proteins 202 to bind the labelled conjugates 205, which then form bound conjugates 222. The bound conjugates 222 still have part of the protein exposed to allow further binding to occur. It should be noted that not all protein 202 molecules may bind with the labelled conjugates 205. Some of the protein 202 molecules may remain in the sampling pad 203 or any other area of the absorbent material during the lateral flow process. It should be further noted that the entire sample 201 may not flow across the entire absorbent material 200. In most examples, a sufficient amount of the sample 201 may remain in the sampling pad 203 after application. Thus the sample 201 may be extracted from the sampling pad 203 at a later stage.

[0075] Fig. 2c illustrates the absorbent material 200 after the lateral flow process. After the lateral flow process, the bound conjugates 222 move through the absorbent material and further bind to the binding bodies 208 on the test area 206. As part of the protein is still exposed in the bound conjugates 222, this allows it to further bind 241 with the binding bodies on the test area. As the bound conjugates 222 bind to the test area 241, the label in the conjugate gives a visual indication of the presence of the conjugate on the test area. As such, the test area 206 gives a visual indication of the presence of the protein 202 in the sample 201. Further, the conjugates that do not bind to a protein 202 flow pass the test area 206 and are able to bind to the binding bodies in the control area 242. These conjugates may be referred to as control conjugates. Similarly to the test area 241, as the conjugates contain a label, the binding of the conjugates at the control area 242 gives a visual indication of this binding. The visual indication of this binding at the control area 242 indicates that the test, as performed by the absorbent material of the sampling device, was performed correctly.

[0076] Method 100 then comprises extracting 102 the sample 201 from the sampling device. In a preferred embodiment, the sample 201 is extracted from the sampling pad 203 of the absorbent material, which forms part of the sampling device. In this sense, the sampling device comprises a sampling pad configured to receive the sample and the method 100 comprises extracting 102 the sample form the sampling pad. Finally, method 100 comprises detecting 103 the protein in the sample using liquid chromatography mass spectrometry (LC-MS). In other examples, the sample is extracted from the wicking pad 210.

[0077] As the sampling device may be configured to provide a visual indication of the protein in the sample, the detection of the protein by the visual indication may be confirmed using the LC-MS analysis. In other words, detecting the protein using the LC-MS analysis comprises confirming detection of the protein as visually indicated by the LFA sampling device. Moreover, as the visual indication provided by the sampling device is captured in the image data, detecting the protein using liquid chromatography mass spectrometry (LC-MS) analysis of the extracted sample can be used to confirm detection of the protein as visually indicated by the image data.

[0078] Fig. 3a illustrates a method 300 for determining a quantity of a protein in a sample. Method 300 may be performed in conjunction with or instead of method 100. Method 300 first comprises capturing 301 image data of a sampling device containing the sample with a camera of a communication device. The communication device may comprise a processor as well as additional modules, that are configured to capture the image data, store the image data on the device and communicate the image data. The processor of the communication device may be additionally configured to perform one or more of the methods described herein.

[0079] The communication device may be any device configured with a camera or image sensor that may capture the image data and communicate the image data to another communication device or to a remote server via the internet or any other means of communication. Examples of a communication device include, but are not limited to, a smartphone, tablet or a digital reader with camera and communication functionally.

[0080] The image data may comprise data indicative of the sampling device and/or the sample. The image data may also comprise additional data including, but not limited to, a GPS location of where the image data was captured and a time stamp including the time and date of when the image data was captured. The sampling device may have a machine readable unique identifier of the sampling device. This machine readable unique identifier may comprise one or more of: a barcode, a QR code or a serial number, that uniquely identifies the sampling device and distinguishes the sampling device from other sampling devices. For example, the machine readable unique identifier may have encoded information that, when decoded, provides information relating to the manufacturer of the sampling device or the protein that the sampling device is designed to detect. The image data captured by the communication device may comprise this machine readable unique identifier, such as a photo of the QR code.

[0081] Similar to method 100, the sampling device comprises an absorbent material and is configured to provide a visual indication of the protein in the sample. The sampling device may also provide an internal standard, which is used in subsequent steps involving LC-MS analysis. The internal standard may be a heavy carbon isotope standard, such as, for example, a polypeptide or protein containing 13C stable isotopes. The internal standard is used in LC-MS analysis to provide a reference standard in the analysis. As the mass-to-charge ratio for the internal standard is known along with the concentration of the internal standard, the internal standard may be used to determine the quantity of the protein in the sample during the LC-MS analysis. This may be performed by comparing the peaks of the internal standard and the peaks of the peptides in the LC-MS spectrum as the location and concentration of the internal standard peak is known beforehand. Rather than comparing peaks, this may be performed by comparing full width at half maximums (FWHMs) or by integrating the area under the peaks and comparing the results of the integration. The comparison to determine the quantity of the protein in the sample may be performed by software, executed by a processor, that is part of a computer system.

[0082] Method 300 then comprises calculating 302 by the communication device, a quantity of the protein in the sample based on the visual indication in the image data. The quantity of the protein calculated based on the visual indication in the image data may be above a determined threshold, such as the limit of the detection (LOD) of sampling device, for example. As described earlier, the visual indication is given by the test area 206 of the absorbent material in the sampling device. The camera may capture the test area 206 and the blank area 211 of the absorbent material in the sampling device. As such, the communication device calculates 302 a quantity of the protein in the sample based on the visual indication in the image data. This step may comprise calculating an intensity ratio in the image data by comparing a first intensity of a test area 206 on the absorbent material with a second intensity of the blank area 211. The blank area 211 corresponds to an area of the absorbent material other than the test area 206. In an example, the second intensity may be determined using the control area 207 as in some scenarios, it is advantageous to take the measurement from the control area 207 instead of the white background of the blank area 211.

[0083] As such, the quantity of the protein in the sample is based on the intensity ratio calculated by the processor of the communication device. That is, the processor calculates the intensity ratio using a mathematical operation using the first and second intensities as input and the intensity ratio as output. For example, the processor calculates the intensity ratio by dividing the first intensity by the second intensity.

[0084] Calculating 302 the intensity ratio may also further comprise creating calibration curves using multiple calibration samples by the processor of the communication device. The calibration samples contain known concentrations of the protein. A calibration curve, also known as a standard curve, is a general method for determining the concentration of a substance in an unknown sample by comparing the unknown to a set of standard samples of known concentration. The calibration curves may be indicative of a linear regression model or a line of best fit.

[0085] The processor of the communication device creates a calibration curve using image data of a sampling device containing a calibration sample containing a known concentration or quantity of the protein, captured by a camera of the communication device. The processor then calculates the intensity ratio in the image data by comparing a first intensity of a test area 206 on the absorbent material with a second intensity of the blank area 211, which corresponds to an area of the absorbent material other than the test area. This intensity ratio with the known concentration is used as a data point to create the calibration curve. This process is repeated using another sampling device containing a calibration sampling containing a known concentration of the protein that is different from the first calibration sample.

[0086] Using the at least two data points determined from the different calibration samples, the processor can calculate the calibration curve. For example, the processor may create the calibration curve using a linear regression model, such as simple linear regression, which provides a linear equation such as y = m x + c. The processor may then apply this linear equation to the intensity ratio calculated from image data to determine an unknown quantity of the protein in the sample.

[0087] In this example, the variable y in the linear equation may be the quantity of the protein in the sample and variable x may be the intensity ratio. The constants m and c are calculated during the process of linear regression and are calculated to minimise the error between the data points determined earlier and the prediction of the quantity based on the linear equation. It is noted that the more data points created, i.e., the more calibration samples used with different concentrations or quantities of the protein, the more accurate the calibration curve will be.

[0088] Adding data points to a calibration curve may be manually added by a user. However, the calibration curves may also be pre-set by the developer of the communication device, enabling the user to only take an image of the sample and use a pre-set calibration curve to estimate the quantity of concentration of the protein in the sample.

[0089] A calibration curve may be created by the processor of the communication device for one particular type of sampling device. In other examples, the processor creates a calibration curve for a group of sampling devices. For example, if the sampling device was a LFA manufactured to detected the presence of a protein relating to a peanut allergen, the processor may create a calibration curve for one particular brand of LFA that detects this protein, or the processor may use a calibration curve for all LFAs that detect this protein. The processor may then store one of the calibration curves in the communication device for use with sampling devices of multiple predefined types.

[0090] Method 300 then comprises sending 303 the image data from the communication device to a server device through the internet or any other means of remote communication of data. The server device may be remotely located from the communication device. The server device may also comprise a processor as well as additional modules, that are configured to store the image data on the server device or further communicate the image data. The processor of the server device may be additionally configured to perform one or more of the methods described herein. For example, the processor of the server device may calculate an intensity ratio of image data, create the one or more calibration curves and determine the quantity of the protein in the sample based on the image data.

[0091] After the server device receives the image data from the communication device, the processor of the server device confirms 304 the identity of the sampling device using the data indicative of the machine readable unique identifier in the image data. The machine readable unique identifier may encode information that, when decoded, confirms the identity of the sampling device. The machine readable unique identifier may also be used to look up information, such as the identity of the sampling device, manufacturer, date/time of manufacture, protein that is detectable by the sampling device, using a look table, such as a database. The machine readable unique identifier may be used to determine whether the sampling device captured in the image data is the same sampling device for which LC-MS analysis is performed on in a subsequent step.

[0092] Finally, in response to confirming 304 the identity of the sampling device, method 300 comprises confirming 305 the quantity of the protein in the sample based on the visual indication in the image data by using the internal standard in liquid chromatography mass spectrometry (LC-MS) analysis of the sample extracted from the sampling device. In this sense, the visual indication of the sampling device may be used as a primarily indication of the presence of the protein in the sample, Subsequently, capturing 301 image data of the sampling device and calculating 302 the quantity of the protein in the sample based on the visual indication in the image data provides a primarily quantification of the protein. The preliminary quantification of the protein is, thus, able to be verified through LC-MS analysis. This is an advantage in situations where an accurate quantification of the protein is desired after a primarily detection and quantification of protein. [0093] Further, the quantity of the protein in the sample may be lower than the limit of detection (LOD) of the sampling device. A quantity that is lower than the LOD would not show a perceivable visual indication of the protein on the sampling device, therefore the protein would be undetected by visual inspection. However, LC-MS analysis can detect quantities of protein lower than the LOD of many sampling devices. Therefore, the LC-MS analysis may additionally detect the protein in the sample from the sampling device.

[0094] The calibration curves used to determine the quantity of a protein in a sample may also be updated using the results of the LC-MS analysis, at the corresponding intensity ratio determined from the image data. In other words, it is possible to update the calibration curves at the intensity ratio in the image data with the quantity of the sample determined using the LC-MS analysis. As a result, many data points would be available to create accurate calibration curves, as these curves can be continually updated. This is an advantage when many communication devices are in communication with the server device, as each communication device sends image data or calculates an intensity ratio to determine a data point for a calibration curve. The data points created from a communication device may be stored on the communication device or communicated to the server device and stored on the server device. The stored data points may be later used to create calibration curves.

[0095] Fig. 3b illustrates a method 350 for detecting a protein in a sample. This method may comprise determining the level of the protein above or below a determined threshold. The determined threshold may be the limit of detection (LOD) of the sampling device, for example. Method 350 may be performed in conjunction with or instead of method 100 or 300. It is noted that method 300 of Fig. 3a and method 350 of Fig. 3b share similarities and even share similar steps. As such, aspects of method 350, such as the communication device, image data, machine -readable unique identifier, server device, LC-MS analysis that were described above for method 300, are equivalent for method 350. [0096] Method 350 first comprises capturing 351 image data of a sampling device containing the sample with a camera of a communication device. This is the same step as step 301 of method 300 in Fig. 3a. Method 350 then comprises processing 352 by the communication device, the image data to detect the visual indication in the image data to thereby detect the protein in the sample using the visual indication in the image data. In some examples, the visual indication of the protein in the sample may not be visible by the naked-eye due to the quantity of the protein being below a certain threshold or due to ambient lightning and other interfering lightning conditions. As such, processing 352 the image may comprise correcting the image data using a correction ratio calculated based on an average of the first intensity and an average of the second intensity. For example, the processor calculates the correction ratio by dividing the average of the first intensity by the average of second intensity. Similar to method 300 of Fig. 3a, the first intensity is indicative of the test area on the absorbent material and second intensity is indicative of an area of the absorbent material other than the test area (i.e., either the control area 207 or the blank area 211).

[0097] As such, correcting the image data may comprise detecting the visual indication of the protein in the sample in the image data to thereby detect 353 the protein in the sample from the image data. Detecting 353 the protein in the sample from the image data provides a preliminary detection of the protein in the sample, which can be later confirmed. It is noted that the visual indication does not necessary mean visual indication by human interpretation. Visual indication may also refer to visual indication by machine interpretation, in such a way, that the visual indication is machine -readable.

[0098] Method 350 then comprises sending 354 the image data from the communication device to a server device through the internet or any other means of remote communication of data. After the server device receives the image data from the communication device, the processor of the server device confirms 355 the identity of the sampling device using the data indicative of the machine readable unique identifier in the image data. Step 354 and 355 of method 350 are equivalent to steps 303 and 304 of method 300. [0099] Finally, in response to confirming 355 the identity of the sampling device, method 350 comprises determining 356 a quantity of the protein in the sample using the internal standard in liquid chromatography mass spectrometry (LC-MS) analysis of the sample extracted from the sampling device. In this sense, detecting 353 the protein in the sample using the visual indication in the image data gives a preliminary indication of the presence of the protein in the sample. This is an advantage in situations where quick detection of the protein in the sample is desired, but the quantity of the protein is not necessarily required, or is not required at the time that the sample is collected using the sampling device.

[0100] Alternatively or additionally, step 356 of method 350 may comprise, in response to confirming 355 the identity of the sampling device, confirming detection of the protein in the sample based on the visual indication in the image data using liquid chromatography mass spectrometry (LC-MS) analysis of the sample extracted from the sampling device. In this sense, the visual indication of the sampling device may be used as a preliminary indication of the presence of the protein in the sample. Subsequently, capturing 351 image data of the sampling, processing 352 image data to detect the visual indication in the image data and detecting 353 the protein in the sample using the visual indication in the image data provides a preliminary detection of the protein. The preliminary detection of the protein is, thus, able to be verified through LC-MS analysis. This is an advantage in situations where an accurate detection of the protein is desired after a preliminary detection of protein. In some example, method 350 may additionally comprise step 302 of method 300. That is, method 350 may further comprise calculating a quantity of the protein in the sample based on the visual indication in the image data. This quantity may also be confirmed by the LC-MS analysis.

[0101] This is also a further advantage as the LC-MS analysis is able to detect the quantity of the protein below or above a threshold, such as, for example, the limit of detection (LOD) of the sampling device or an acceptable limit of protein for food safety. For example, if sampling device or the image data does not provide a visual indication of the protein in the sample, then this result can be confirmed by the LC-MS analysis as it can determine the quantity of the protein below the LOD of the sampling device. Confirming the result by LC-MS analysis may also help improve the detection of the protein in the sample by improving the processing, by the communication device, of the image data in step 352. For example, the results of the LC-MS analysis may be used to update the calibration curves used by the communication device.

[0102] Any one of the previously described methods may further comprise determining peptide sequence information of the proteins captured by the sampling device using the LC-MS analysis. After collecting the sample with the sampling device, the sample is extracted from the sampling device with an extraction solution. Then a digestion solution may be ran through the sample. The digestion solution is used to digest the protein(s) in the sample and into their constitute peptides. The solution with the constitute peptides may then travel through the liquid chromatography (LC) system, which separates the peptide in the mixture according to their adherence to a solid adsorbent material within the column of the LC system. Thus, the peptides come out of the column individually, which can individually be analysed by the mass spectrometer. This provides the identity of the peptides present in the sample, and may provide the peptide sequence information. The peptide sequence information and or the information of the mass to charge ratio of the peptide and it’s fragments (in MS/MS mode) can then be used to identify and/or quantify the protein(s) that were present in the sample.

[0103] Any one of the previously described methods may also further comprise detecting the presence of a second or additional proteins using the LC-MS analysis to verify a purported origin of the sample. For example, the sample may contain one or more proteins in addition to the protein of interest. The detection or lack of detection of expected proteins in the sample may confirm where the sample has been collected from. For example, if one wants to determine the presence of peanut traces in tomato soup, the method may further comprise detecting a secondary protein, such as a tomato protein. Detecting the presence of the tomato protein would indicate that the sample has be collected from tomato soup. Conversely, if a user transports, delivers or sends a sampling device that contains a sample and claims the sample was collected from tomato soup, this claim can be verified by detecting the presence of a tomato protein in the sampling device via LC-MS analysis.

[0104] Fig. 4 illustrates an example system 400 for performing one or more of the methods described herein. For example, system 400 may be used to perform method 100 illustrated in Fig. 1. In another example, system 400 may be used to perform method 300 illustrated in Fig. 3a. In yet another example, system 400 may be used to perform method 350 illustrated in Fig. 3b. For explanatory purposes, references to method 300 and method 350, as well as the steps of method 300 and the steps of method 350 are used to describe system 400, however system 400 is not limited to performing method 300 or method 350. System 400 comprises a sampling device 401 and a communication device 402. Communication device 402 may comprise a processor 403 connected to a program memory 404 and a data memory 405. The program memory 404 is a non-transitory computer readable medium, such as a hard drive, a solid state disk or CD-ROM.

[0105] Software, that is, an executable program stored on program memory 404 causes the processor 403 to perform some steps of the method in Fig. 3a, that is, processor 403 captures 301 image data of the sampling device, calculates 302 a quantity of the protein in the same based on the visual indication in the image and sends 303 the image data to a server device. Software may also perform some of the steps of method 350 in Fig. 3b, that is, processor 403 captures 351 image data of the sampling device, process 352 image data to detect the visual indication in the image data, detects 353 the protein in the sample using the visual indication in the image data and sends 354 the image data to a server device.

[0106] The data memory 405 may store image data and retrieve the image data for later use. The data memory 405 may also store data relating to the quantity of the protein in the sample, as well as data relating to the calculation of this quantity including, but not limited to, the intensity ratio of image data, the correction ratio of image data, data points created from image data of calibration samples and calibration curves. The image data may also be indicative of a two-dimensional image, which may be stored on data memory 405 as Joint Photographic Experts Group (JPEG) format, RAW image format or a similar/equivalent image format, for example. The processor 403 may receive data through all these interfaces, which includes memory access of volatile memory, such as cache or RAM, or non-volatile memory, such as an optical disk drive, hard disk drive, storage server or cloud storage.

[0107] The communication device 402 further comprises a camera 406 that is in communication with the processor 403 via the input/output (I/O) port 407. In some embodiments, camera 406 is integrally contained in the communication device 402, despite being drawn as disjointed entities in Fig. 4. Camera 406 captures image data of the sampling device 401, which is then communicated to the processor 403 via the I/O port 407.

[0108] It is to be understood that any receiving step may be preceded by the processor 403 determining or computing the data that is later received. For example, processor 403 may then store the image data on data memory 405, such as on RAM or a processor register. Processor 403 then requests the data from the data memory 405, such as by providing a read signal together with a memory address. The data memory 405 provides the data as a voltage signal on a physical bit line and the processor 403 receives the image data as the input image via a memory interface.

[0109] Steps 301-303 of Fig. 3a and steps 351-354 of Fig. 3b are to be understood as a blueprint for the software program and may be implemented step-by-step, such that each step is represented by a function in a programming language, such as C++ or Java. The resulting source code is then compiled and stored as computer executable instructions on program memory 404.

[0110] It is noted that for most humans performing the method 300 or method 350 manually, that is, without the help of a computer, would be practically impossible. Therefore, the use of a computer is part of the substance of the invention and allows performing the necessary calculations that would otherwise not be possible due to the large amount of data and the large number of calculations that are involved. [0111] The communication device 402 may further comprise a communication port 408, allowing the communication device 402 to establish a communication with a server device 409. The communication port 408 may have the same input and output functionally as I/O port 407. Thus, in some example, the I/O port 407 and communication port 408 may be the same port, although they are drawn as distinct entities in Fig. 4. In another example, communication port 408 may be a wireless transmitter and/or receiver, that is configured to send and receive data, such as the image data, with the server device 409. Similar to communication device 402, server device 409 may comprise a processor, data memory, program memory, and one or more I/O or communication ports. In this sense, server device 409 may have similar functionally to the communication device 402, and may be configured to perform similar methods and/or processes to that of the communication device 402.

[0112] While only one communication device 402 is depicted in Fig. 4, there may be a network of many communication devices that may communicate with the server device 409. The network of communication devices may also communicate directly with one another via the internet, any other means of wireless communication or wired connections. In an example, the same calibration curve may be communicated, over a communication network, from the server device 409 to the multiple communication devices. Each of the multiple communication devices may then use the calibration curve to determine the quantity of a protein in a sample from image data that each of the multiple communication devices capture.

[0113] The server device 409 may retrieve and store data such as, but not limited to, image data, calibration curves and spectrum produced during LC-MS analysis. Additionally, the server device 409 may then communicate this data back to the communication device 402 and/or a computer 410. The image data may be received from a source external to system 400, such as another system located remotely to system 400, which may be in communication with the server device 409. The remote system may be located within the same facility as system 400 or may be completely remote from system 400. The image data may be stored in the server device 409 as Joint Photographic Experts Group (JPEG) format, RAW image format or a similar/equivalent image format. Similarly, processor 403 receives the image data via the input/output port 308 and performs aspects of method 300 of method 350 on the image data. After processor 403 performs aspects of method 300 or method 350 on the image data, the image data, the quantity of the protein in the sample and/or the detection of the protein in the sample determined by processor 403 may be communicated to the server device 409. The server device 409 may then communicate the image data and/or the quantity of the protein in the sample to an external system or store them.

[0114] Server device 409 may be configured to create and store calibration curves used to calculate the quantity and detect the protein above a set threshold level in the sample using the visual indication of the protein in image data. For example, the communication device 402 may capture multiple instances of image data of sampling devices with different calibration samples and communicate the multiple instances of image data to server device 409. Server device 409 may then calculate an intensity ratio for each instance of image data by comparing a first intensity of a test area on the absorbent material with a second intensity of an area of the absorbent material other than the test area. Each intensity ratio and the known concentration of the protein in the calibration samples can then be used by the server device 409 to create the one or more calibration curves. These calibration curves can then be stored on the server device 409 and later retrieved and/or communicated to the communication device 402 to calculate the unknown quantity of a protein in a sample.

[0115] System 400 may also comprise a computer 410 that may be in communication with a liquid chromatography mass spectrometry (LC-MS) system 411. In some examples, the computer 410 and the LC-MS system 411 may be a single entity or may be multiple entities, as depicted in Fig. 4. Computer 410 may comprise a processor, a data memory module, a program memory module and one or more I/O or communication ports, similar to the communication device 402. Computer 410 may be configured to provide instructions and/or control the operation of the LC-MS system 411 and may be further configured to retrieve and/or receive data from the LC-MS system 411. Computer 410 may then process the data communicated from the LC-MS system 411, which may include creating a mass spectrum of the sample extracted from the sampling device and/or determining the quantity of the protein in the sample based on the visual indication in the image data.

[0116] In system 400, a sample may be collected from a first location using the sampling device 401 and the LC-MS analysis may be performed by the LC-MS system 411 at a second location remote from the first location. In this situation, the sampling device 401 that contains the sample collected from the first location may be transported from the first location to the second location, as indicated by path 412. In this sense, the sampling device 401 may be used as a sample carrier, but may also be used as a preliminary means of detection of the protein in the sample, due to the visual indication provided by the sampling device 401. This may be also be referred to as “on-site testing”. Thus, the LC-MS system 411 may be used to confirm the detection of the protein in the sample based on the visual indication, which may be referred to as “confirmatory testing”. Further, when integrated with the communication device 402 and the server device 409, the communication device 402 may calculate a quantity of the protein in the sample based on the visual indication. Thus, the communication device 402 may provide a preliminary quantity/detection of the protein in the sample. The LC-MS system 411 may then analyse the sample by extracting it from the sampling device 401 and the results from the analysis may be communicated to computer 410. Computer 410 may then calculate the quantity of the protein in the sample based on the LC-MS analysis, thereby confirming the preliminary quantity/detection of the protein in the sample based on the visual indication.

[0117] Computer 410 may then communicate the results of the LC-MS analysis, as performed by the LC-MS system 411, to the server device 409. The server device 409 may then communicate these results to the communication device 402. As such, the communication device 402 may display the quantity of the protein calculated by processor 403, as well as the quantity determined by the LC-MS system 411 by LC-MS analysis. [0118] Software may provide a user interface presented to the user on communication device 402 and/or computer 410. The user interface is configured to accept input (via buttons or text fields etc.) from the user, via a touch screen or a device attached to communication device 402 and/or computer 410 such as a keyboard or computer mouse. These devices may also include a touchpad, an externally connected touchscreen, a joystick, a button, and a dial. In an example, communication device 402 and/or computer 410 may display multiple instances of image data, and the user may choose one of the multiple instances of image data for method 300 to be performed on, by processor 403. The user may choose one of the multiple instances of image data by interacting with the touch screen or inputting the selection with a keyboard or computer mouse.

[0119] The processor 403 may receive or send data, such as image data, from data memory 405 as well as from the I/O port 407 and/or the communication port 408. In one example, the processor 403 sends image data from the communication device 402 to server device 409 via communications port 408, such as by using a Wi-Fi network according to IEEE 802. 11. The Wi-Fi network may be a decentralised ad-hoc network, such that no dedicated management infrastructure, such as a router, is required or a centralised network with a router or access point managing the network. Communication between server device 409 and computer 410 may similarly established using a Wi-Fi network according to IEEE 802.11. System 400 may further be implemented within a cloud computing environment, such as a managed group of interconnected servers hosting a dynamic number of virtual machines.

[0120] Although communication port 408 is shown as single entity, it is to be understood that any kind of data port may be used to receive data, such as a network connection, a memory interface, a pin of the chip package of processor 403, or logical ports, such as IP sockets or parameters of functions stored on program memory 404 and executed by processor 403. The parameters of functions may be stored on data memory 405 and may be handled by-value or by-reference, that is, as a pointer, in the source code. [0121] Fig. 4 is one example of a configuration of system 400. However, system 400 is not strictly limited to this configuration and this may be one possible embodiment of system 400. In another embodiment, computer 410 may be in direct communication with the communication device 402.

[0122] The following sections provide methods for testing the method described above. In this disclosure, three commercially available lateral flow assays, two for peanut and one for gluten detection, were used as exemplary screening tests for which such assays can be developed.

Methods

[0123] Ratio-metric ambient light correction optimisation

[0124] Ratio-metric approaches and shutter time optimisation were conducted using gold nanoparticles (GNPs) dropped onto filter paper at various concentrations. GNPs were synthesised following the Turkevich method. Calibration curves were photographed as both RAW and JPEG images using an Android (Galaxy S21 Ultra) with the white balance fixed at 5000, ISO at 100 using 1/30, 1/45 and 1/60 shutter times and backlit conditions. All photographs were backlit to reduce illumination variation. From this, ImageJ (VI.53k, https://imagej.nih.gov/ij/) was used to extract RGB values from JPEG files and intensity values from RAW files. Readings were taken from the test area, the control area and the blank area, corresponding to the white section of the absorbent material just below the test area. These measures were used to compare the performance of both uncorrected and ratio corrected values using either the control area or the white background as the denominator. The smartphone cameras were operated in Pro-mode (Android) or using the ProCam application for the iPhone (model S6) to enable full control over the camera settings. The shutter times 1/60, 1/45 and 1/30 were tried in optimisation experiments.

[0125] Peanut and gluten materials [0126] A 99.9% pure commercial peanut butter (CPB) from roasted Australian peanuts was defatted and used as a standard for peanut allergen following reference. For defatting, the CPB was vortexed for five minutes and sonicated in n-pentane at a 1: 10 (w/v) sample/n-pentane ratio. Subsequently, samples were centrifuged for 10 minutes, and the resultant supernatant removed. This step was repeated once more. Rotary vacuum concentration (15 min; 40°C) was used to dry defatted samples. Grain samples from the wheat cultivar Chara were obtained from the Australian pasture collection and Australian Winter Cereals Collection (Tam worth, Australia). The samples were manually inspected to exclude any foreign seed contaminations. Flour samples were obtained by milling the grains with a Metefem Hungarian Mill (model FQD2000, Hungary).

[0127] Commercial Lateral Flow Assay (LFA) analysis

[0128] Peanut calibration curves were constructed by serial 2-fold dilutions from 2000 pg mL-1 to 0.49 pg mL-1 in phosphate buffered saline (PBS; pH 7.2). For gluten, a calibration curve was constructed by serial 2-fold dilutions from 8000 pg mL-1 to 0.98 pg mL-1 in PBS. The LFA assay protocols provided by the manufacturers (Romer and 3M) were followed for all experiments. The results were photographed using the fixed camera settings that were previously optimised using the GNP spots.

[0129] LFA extractions and LC-MS analysis

[0130] Immediately after being photographed, the sampling pad and wicking pad, as well as the test area from each assay, were carefully sliced from the nitrocellulose membrane using a scalpel and put onto separate 0.45 pM spin filters. A scalpel was used to pull apart the fibres of the wicking or sampling pad to facilitate extraction. Samples were kept frozen until analysis. Subsequently, 400 pL of extraction solution (8 M Urea, 50 mM dithiothreitol) was added to each spin column and sonicated for 10 minutes, then transferred shaken (600 RPM) at ambient temperature for 30 minutes before being centrifuged at 14,000 RCF for 5 minutes. The supernatant was transferred to 10 kDa Amicon filters. These were spun for 15 minutes at 14,000 RCF. Each column was washed thrice using a solution of 200 pL of 50 mM Ammonium bicarbonate, 1 mM calcium chloride solution (ABC) at 14,000 RCF for 10 minutes. After this, the collection tube of each sample was changed, discarding flow-through. A 200 pL aliquot of trypsin solution (5 pg ml-1) was added and extracts were digested for 1 hour at 37°C on a heating block at 600 RPM. Next, filters were spun down for 15 minutes in a centrifuge at 14,000 RCF and the flowthrough was collected. Then 200 pL ABC was added and filters were centrifuged for 15 minutes at 14,000 RCF. Flowthrough of the wash and collection steps were pooled and lyophilised in a speed vac. Lyophilised peptides were solubilised in 50 pL of acidified water spiked with iRT (0.1% FA, 0.05 iRT picomol pL-1) and transferred to LC vials and stored in a freezer pre-analysis. A scheduled multiple reaction monitoring (sMRM) method for gluten and peanut quantification was followed for the analysis of the sample extracts. The samples (8 pL per injection) were analysed on a SCIEX 6500 QTRAP mass spectrometer coupled to a Shimadzu Nexera UHPLC system.

[0131] Smartphone application development

[0132] An Android app has been built to take images of the absorbent material. The app was built using the Android's Camera2 API (https://developer.android.com/training/camera2) source code. The macro camera was selected to gain better focus for the close proximity targets. The developed app allows the user to zoom into the test and the control area and then move the two detection zones manually to select the test-strip areas, such as the test area and the control area, that need to be read to calculate a ratiometric value and the average colour/intensity result is displayed on the screen. The ratiometric value can then be interpolated in the app using a calibration curve function that was created using a set of LFAs photographed at known concentrations. The ISO, shutter time and white balance can be fixed in the app. Images can be saved as JPEG or DNG (RAW) images.

[0133] The smartphone application (app) is software that, when installed on a communication device, such as an Android smartphone, and executed by the communication device, causes the communication device to, for example, capture image data of the sampling device with a camera; calculate an intensity ratio in the image data by comparing a first intensity of a test area on the absorbent material with a second intensity of an area of the absorbent material other than the test area; wherein the test area provides the visual indication of the protein in the sample; and calculate a quantity of the protein present in the sample based on the intensity ratio.

[0134] Software and statistics

[0135] ImageJ VI.53k was used to extract RGB and RAW intensity values from captured images. To extract RAW data, data from photos were linearised using the using the DCRaw reader plugin (v. 9.27x ) in “totally RAW” document mode with a 16-bit output. Spectra analysis and method development were conducted using Skyline v21.1.0.278. GraphPad 8.0.1 and Inkscape 1.1.1 were used to prepare and assemble statistical analysis and figures.

Results

[0136] Shutter-time optimisation and preliminary evaluation of image format colour channels

[0137] A calibration curve of various GNP concentrations (done in triplicate) was dropped on filter paper and photographed using an Android phone in both RAW and JPEG image formats to optimise the camera settings and test the performance of the ratio metric ambient light correction. Shutter times of 1/30, 1/45 and 1/60 were used with all other camera settings kept constant. R, G, B, average grayscale, and weighted grayscale measurements were extracted from 8-bit JPEG images and intensity values were extracted from 16-bit RAW images. These values were fitted using a four- parameter log(dose)-response curve.

[0138] Fig. 5 illustrates different calibration curves for relative gold nanoparticle (GNP) concentration created using image data from a communication device. Graphs A-B show calibration curves for relative GNP concentrations from uncorrected intensity values of JPEG colour channels (red, blue, green), average grayscale and weighted grayscale values (A) and raw intensity values (B). Graphs C-G show calibration curves for relative GNP concentration values from ratio (signal/white background) corrected JPEG colour and greyscale channels. Graph H shows calibration curves constructed for relative GNP concentrations from ratio corrected raw intensity values. All images (n=3) were taken with an Android in pro-mode with fixed ISO, white balance and zoom settings. Shutter times were varied (1/30, 1/45, 1/60) as indicated in legend.

[0139] The summary statistics (R2, coefficient of determination; LOD, limit of detection; IC50, the concentration of the target where the sensor response is reduced by half) of the curve fits are shown in Table 1.

[0140] Table 1. Summary of the fit of calibration curves shown in Fig. 5.

[0141] Ratio corrected values, where the average extracted intensity of the test area was divided by the average extracted intensity of the blank area of the absorbent material (between the test and control area), were mainly shown to have similar or slightly lower LOD and IC50 values. In this sense, the method for detecting the protein in the sample and/or determining the quantity of the protein in the sample further comprises correcting the image data using a correction ratio calculated based on an average of the first intensity and an average of the second intensity. The slightly lower LOD and IC50 values may be caused by a reduction in background illumination noise.

[0142] The correction ratio may also be used to qualitatively detect the protein in the sample from the LFA (sampling device). Generally, detecting the protein using the LFA relies on the visual indication, which is read by the naked-eye. However, as illumination noise and other lighting factor may affect the ability for a user to determine the presence of the visual indication, the protein may not always be detected by a naked-eye reading. Therefore, by capturing image data of the sampling device containing the sample, the communication device that captures the image data may calculate a correction ratio and correct the image data. The communication device may then be able to detect the visual indication of the sampling device, or a user may be able to see the visual indication after the correction. Therefore, any one of the method described herein may further comprises correcting the image data using a correction ratio and detecting the protein in the sample from the corrected image data.

[0143] The red colour channel underperformed. This may be because varying red background noise particularly disturbed the measurements whereas greyscale or blue channel values were less affected by this artefact in the images. However, ratio correction did have a negative effect on the fit (R2) of the calibration curve to the data across JPEG and RAW image formats. This negative effect was the least prominent in the RAW ratio corrected images. [0144] A very notable finding is that ratio correction mitigates the impact of shutter time, as a proxy for light intensity, on the yielded intensity values as shown by the corrected ratio calibration curves falling over each other across the investigated shutter times (Fig. 5C-H) in comparison with the uncorrected calibration curves (Fig. 5A-B). Although this was observed to varying extents across the colour channels within JPEG images, this effect was most distinct in the ratio corrected RAW images (Fig. 5H). This suggests that raw images enable high quality corrections for light intensity changes within this straightforward ratio metric approach.

[0145] Being able to correct for light intensity variations is useful for enabling consistent in-field interpretation of colour intensity changes in colorimetric tests. In addition to this, LODs and IC50 values were comparable, or lower as the colour and greyscale LODs and IC50 values for all trialled shutter times while the R2 values were superior. The 1/60 shutter time setting produced comparable R2 values for all trialled colour channel, greyscale, and RAW images and better LOD values across the board suggesting that this shutter time enabled the highest sensitivity. This is likely as the shorter shutter time was exposed to less light, meaning environmental light conditions would have less of an impact on the end image produced; this shutter time was chosen from here on. Additionally, only ratio metric values were used from here on since this method showed excellent robustness against ambient light variations, which are unavoidable in the field.

[0146] Comparison of image format and ratio correction across mobile devices

[0147] Fig. 6 illustrates calibration curves of GNP concentration and intensity in colour channels of JPEG and RAW images captured using Android and iPhone mobile phone models at 1/60 shutter time. Inter-phone variation has an effect on LFA quantification using ratio-metric measurements for the various tested colour channels and intensity measurements. The ratio-metric approach improves calibration curve overlap compared to uncorrected calibration curves. Clear differences in the amount of calibration curve overlap and shape can be observed for the tested image and colour/intensity formats applied. Ratio-corrected raw intensity and blue channel JPEG calibration curves enabled the closest fits between the two phone models compared. Notably, ratio corrections decreased the R2 of the Android calibration curves but improved or did not affect the fit of the iPhone calibration curves compared to the R2 values obtained for the uncorrected data in the JPEG image format. This shows that ratio correction varied in effectiveness depending on the phone model when the JPEG format was used. For the raw images, R2 values were excellent for both phone models and did not differ between uncorrected and ratio corrected calibration curves.

[0148] The raw intensity values also led to particularly low relative standard deviations (RSDs) for both phones (iPhone 1.05 ± 0.51%; Android 1.41 ±0.42%); was the only measurement type that enabled the detection of GNPs at a concentration below a relative value of one; and, was the only measurement type that showed no plateau in the higher concentration range of the GNPs for both phone models. As such ratio corrected raw values were used from here on as these were considered to outperform all JPEG colour channel and intensity measurements in terms of ambient-light variation robustness, inter-phone variation, dynamic range, precision and sensitivity. Other techniques aiming to overcome this limitation such as the use of a colour chart added to each image or a camera calibration approach can be incorporated to further reduce the issue of inter-phone variation.

[0149] Validation of image quantification using allergen LFA test strips

[0150] Calibration curves were constructed using ratio metric raw intensity values for two commercial peanut LFAs (Romer and 3M) and a gluten LFA (3M) using the Android and the iPhone (Fig. 7; Table 2). The iPhone underperformed compared to the android in terms of R2 values and the A ratio observed between the maximum and minimum signal across all investigated assays. Nonetheless, high R2 values (0.94<R2<1.00) and exceptionally low intra-day repeatability Relative Standard Deviations (RSDs;between 1.0 and 7.0%) were obtained for all calibration curves.

[0151] Table 2: Analytical parameters of the calibration curves shown in Fig. 7.

[0152] Overall, the LODs and linear ranges for the iPhone slightly outperformed (maximum ~2-fold improvement) compared to the Android values. Sensitivity of the peanut assays was good varying between 3.4 (iPhone) and 6.8 (Android) ppm. These values are above the reported LODs of these assays by the manufacturer (1 ppm) but remain in the same order of magnitude. Notably, no clear test area coloration was observed upon visual inspection of the test lines at peanut levels below ~5 ppm for both the Romer and 3M peanut assays confirming that visual and smartphone derived LODs were similar for these assays for both tested models.

[0153] Fig. 7 illustrates smartphone based quantification of commercial lateral flow assays. Graph A shows calibration curves (n=3) for the Romer and 3M peanut LFA assays. Graph B shows calibration curves (n=3) for the 3M gluten LFA assay. The iPhone calibration curves of the Romer and 3M assays show excellent overlap while the Android calibration curves do not (Fig. 7A). This indicates that assay specific calibration do not need to be corrected for commercial assay variation for certain phone models with this ratio-metric approach. The inter-day repeatability was also tested for the peanut LFA assays with new samples tested one month after the calibration curve generation. This test was performed at 12 and 4 ppm for the Romer assay and 10 and 5 ppm for the 3M assay with the android and iPhones (n=3). The average error on the predictions was 38% and 37% for the iPhone Romer and 3M assays, respectively. The average error on the predictions was 13% and 47% for the Android Romer and 3M assays, respectively. The Android-based quantification of the Romer assay did produce excellent results in this final test. Moreover, the assays enable peanut determination over a more than two orders of magnitude long range. As such even a 50% (or 2-fold) prediction error may still be acceptable, especially as a semi-quantitative result that can be followed up with LC-MS based quantification.

[0154] Smartphone application functionality

[0155] The developed Android app allows users to take images of test strips at the determined optimal camera settings and calculate average colour/intensity values from JPEG and raw images using the RGB and greyscale colour/intensity algorithms. The smartphone can be placed directly above the image as long as there is an angled light source that eliminates shade formation on the images.

[0156] Fig. 8 illustrates a smartphone application for colour quantification. Captured images can be processed for quantification and the data can be stored in the cloud for later use. Fig. 8 shows the lateral flow assay prior to zooming in on the test line. The red box indicates the zoom area. The right image shows the zoomed in area with the boxes used to select the test strip area and white area used for ratio metric value calculation. Average colour area values are indicated.

[0157] The captured images can be stored in a cloud database and can also be processed for quantification and stored for later use. Two individual blocks can be dragged by the user to the relevant test strip/background areas and the average colour results are displayed on the screen (Fig. 8). The Android app can also analyse images taken from other devices. Colour analysis of the images may also be adapted using the DCraw algorithm as used in the ImageJ plugin, given the discovered substantial benefit of using raw intensity values.

[0158] The RSD values of the next best channel values (the blue channel) determined by the app were determined for the Romer Peanut LFA dataset and were 2.4 ± 2.0 %. This is significantly lower (p=0.006; two-tailed t-test) than the RSD values for the manually annotated raw intensity values (9.5 ±4.6%) showing the potential of automatic annotation. Manual annotation may also be performed using the smartphone application for LFA assay quantification. [0159] Optimisation of LC-MS extraction protocol and direct quantification of allergenic peptides from LFAs

[0160] Fig. 9 illustrates optimisation of the LC-MS extraction protocol. Graph A shows the average peak area values (n=3). Graph B shows the average coefficient of variation (%CV or simply CV) for sampling and wicking pad extracts of the top ten identified peanut peptides. Peanut extracts from the sampling pad of multiple sampling devices, wicking pads and test area of the LFAs were analysed by LC-MS to determine which section of the LFA is best suited for LC-MS analysis. The top ten best performing peptides were then selected for further assay optimisation. For this selection a cut-off was used whereby the signal intensity of the wicking pad and/or the sampling pad needed to have an ion count of at least 1* 10 ⁶ a.u. The peak area obtained for sampling pad extracts was considerably higher than the peak areas obtained for the wicking pad for all selected peptides (Fig. 9A). Thus, the sampling pads likely contained the highest peanut protein concentration across the multiple different sampling devices. The peak area CVs for the sampling pads were also considerably lower as the wicking pad CVs for nine of the 10 peptides tested with various CVs well below 10% (Fig. 9B). Sampling pad CV’s may have been best performing as the composition of protein extracts from wicking pads were more sensitive to unintentional variation in the extraction and LFA assay protocol. Such variation would cause larger changes to the CVs in the wicking pad extracts due to the lower protein concentration in these extracts. Thus, sampling pads were used for the remainder of extractions.

[0161] LC-MS assay performance

[0162] LC-MS performance was tested for the quantification of peanut and gluten protein directly from the sampling pads of the commercial LFAs with the quantitative transitions of the best performing peptides for each assay (TANDLNLLILR for peanut and LEGSDALSTR for gluten determination). It is noted that each letter of the peptide represents one of the 20 common amino acids found in proteins. [0163] Fig. 10 illustrates the direct quantification of allergenic peptides from LFAs by LC-MS. Graph A shows a calibration curve (n=3) constructed from sample pad extracts from the Romer peanut assay. Graph B shows a calibration curve (n=3) constructed from sample pad extracts from the 3M peanut assay. Graph C shows calibration curve (n=3) constructed from sample pad extracts from the 3M Gluten assay. Inset figures show the peaks of the quantitative and qualitative transitions at the LOD. Graph D shows RSDs (n>18) calculated forthe calibration curves shown in Figs. 10A-C.

[0164] An excellent linear range was obtained for all three calibration curves ranging from 2000 to ~8 ppm for both peanut LFA types and 4000 to ~2 ppm for the gluten LFAs with R2>0.95 forthe peanut and R2>0.93 for the gluten assays (Fig. 10A-C). The limit of quantification (LOQ) for all assays were quite close to one another (7-8 ppm). Notably forthe peanut LFAs the LC-MS LOQs were slightly above the LOD of the smartphone LOD while for gluten the LC-MS LOQ was far below the smartphone LOD. All three LC-MS assays had LODs well below the LODs determined for the smartphone assays as well as the visual LODs reported by the manufacturer.

[0165] Regarding repeatability, the LC-MS quantification of the Romer peanut LFAs performed well (CV = 8.5 ± 6.3%) shortly followed by the 3M assay (CV = 15.2 ± 4.4%; Fig. 10D). The gluten assay performance was slightly lower in terms of repeatability (CV = 28± 8.0%). As such, the linear range, LOD, LOQ and RSD parameters of the LC-MS assays for peanut quantification directly from LFAs all comply with the AOAC Standard Method Performance Requirements (SMPRs®) for Detection and Quantitation of Selected Food Allergens (AOAC SMPR 2016.002).

[0166] For gluten determination the LOQ and CV are slightly above the recommended limit in the AOAC SMPR® for Quantitation of Wheat, Rye, and Barley and could use further optimisation. However, the LC-MS assays are clearly able to detect false negatives from LFAs due to antibody instability and matrix affects since distinctive peaks can be distinguished well below the AOAC recommended LODs (Fig. 9A-C). Moreover, a protein BLAST search shows that LEGSDALSTR is a unique identifier for wheat gluten (Triticum spelta, T. aestivum and T. dicoccoide) except for one hit for a hypothetical E. col protein (WP_161424595.1; NCBI) while TANDLNLLILR is a unique identifier for cultivated peanut and wild peanut protein (both of which carry the peanut allergens). Thus, false positives caused by antibody cross reactivity can be identified with this method.

[0167] Moreover, untargeted proteomic strategies can be applied on the extract if no target peptides are detected by this targeted LC-MRM-MS assay to determine which proteins are present in the sample. This may help to identify potential candidates in the food matrix causing cross reactivity or identify the presence of unexpected proteins or absence of any protein, which may indicate the test was performed incorrectly. As such, any one of the method described herein may additional comprise detecting a false negative or a false positive by the sampling device (LFA) using the LC-MS analysis to detect the protein below a limit of detection of the sampling device.

Outcomes

[0168] Smartphone-based quantification

[0169] The results presented in this disclosure clearly show that 16-bit raw intensity values from DNG images produce superior results for smartphone -based colour quantification when compared with 8-bit RGB channels and grey-scale values obtained from JPEG images and reached similar LODs as observed with the naked eye, which is an improvement to previous smartphone based quantification. This is likely because the raw image format remains unprocessed, causing less artefacts, and has an improved dynamic range.

[0170] Ratio-metric ambient light corrections also performed particularly well for the raw image format, and it was shown that such corrections can be successfully used to avoid the use of additional light sources or a light-shielding box. Instead, a simple picture with fixed camera parameters can be taken and used for the robust quantification of test-strips. Moreover, striking calibration curve overlap was observed for two different LFA assays for peanut quantification when the same phone model was used suggesting it may be possible to construct general calibration curves that can be applicable for multiple commercial assays detecting the same target if a given phone model is used.

[0171] LC-MS based verification of LFA results

[0172] Direct quantification from commercial LFAs via LC-MS showed great promise with excellent analytical performance observed for both peanut assays and acceptable performance for the gluten assay. The method can detect false negative and false positive results as it enables detection well below the LOD of the LFAs and enables unambiguous confirmation that the target proteins were in the sample. As such it may proof very valuable for LFA result conformation as well as LFA development and performance characterisation.

[0173] It will be appreciated by persons skilled in the art that numerous variations and/or modifications may be made to the above-described embodiments, without departing from the broad general scope of the present disclosure. The present embodiments are, therefore, to be considered in all respects as illustrative and not restrictive.