CROSS-REFERENCE

[0001] This application claims priority to U.S. Provisional Application No. 62/645,967, filed March 21, 2018, which is entirely incorporated herein by reference.

BACKGROUND

[0002] Profiling of microbial compositions is critical for assessing health of an individual.

For example, the human gut microbiome plays a key role in many health conditions, including obesity, gastrointestinal health, nutrient absorption, and drug metabolism among others. Despite this understanding of the interrelation between microbiomes and health, the complexities of the microbiomes, cost of such assessment, and a long sample-to-answer duration have made regular monitoring of microbiomes challenging.

[0003] Metagenomic approaches to understanding the microbiome can further illuminate the roles of microbiomes and have only recently been enabled by“next-generation” sequencing technologies. While the information uncovered by these studies will become increasingly valuable to those interested in targeting the microbiome for therapeutic interventions and consumer products, using these technologies remain expensive due to expensive instruments and complex analysis. Further, these technologies are not accessible for regular monitoring in a point-of-need setting, such as at home.

[0004] The present disclosure provides solutions to these limitations by providing devices, systems, and methods for taxonomic identification of microorganisms and functional identification of genes in microbiomes that is simple to use in a point-of-need setting with a short sample-to-answer time. Further, the disclosure provides an inexpensive and simple solution for regular monitoring of the microbiome. Furthermore, these systems can be used for metagenomic analysis of nucleic acids from unprocessed samples, such as stool sample.

SUMMARY

[0005] The present disclosure provides devices, systems, and methods for profiling of microbial compositions in a point-of need setting with a short sample-to-answer time, thereby providing a simple solution for assessing the microbiome of a sample. In an aspect, the present disclosure provides a fluidic cartridge for performing nucleic acid analysis of fecal- derived microbiome, wherein the fluidic cartridge is capable of processing a fecal sample and performing nucleic acid analysis of fecal-derived microbiome. [0006] In some embodiments, the cartridge comprises a plurality of fluidic processing elements prefilled with reagents for processing the fecal sample to extract and label nucleic acid molecules, and a microarray comprising probes capable of hybridizing with the labeled nucleic acid molecules to perform nucleic acid analysis.

[0007] In some embodiments, the fluidic cartridge comprises a sealable opening for depositing the fecal sample. In some embodiments, the sealable opening is compatible with a sample collector. In some embodiments, the fluidic cartridge further comprises a vibratory motor for homogenizing the fecal sample. In some embodiments, the fluidic cartridge further comprises an ultrasonic transducer for homogenizing the fecal sample. In some embodiments, the extracted nucleic acid molecules are not amplified using polymerase chain reaction.

[0008] In some embodiments, the fluidic cartridge comprises a filter through which the fecal sample is passed through to extract nucleic acid molecules. In some embodiments, the filter is a monolithic porous plug with a defined pore size. In some embodiments, the defined median pore size is about 1 nanometer to 200 micrometer in diameter. In some embodiments, the filter is chemically treated. In some embodiments, the fluidic cartridge further comprises a plurality of filters. In some embodiments, the fluidic cartridge further comprises at least 2 filters. In some embodiments, the fluidic cartridge further comprises at least 3 filters. In some embodiments, at least one filter of the plurality of filters is configured to remove enzymatic inhibitors. In some embodiments, at least one filter of the plurality of filters is configured to grind the fecal sample. In some embodiments, at least one filter of the plurality of filters is configured to capture nucleic acid molecules for the nucleic acid analysis. In some embodiments, the filter is configured to retain beads. In some embodiments, the beads are coated with a chemical moiety for removing enzymatic inhibitors. In some embodiments, the beads are capable of grinding the fecal sample. In some embodiments, the beads are coated with a chemical moiety for capturing nucleic acid molecules.

[0009] In some embodiments, the fluidic cartridge further comprises a pump, wherein the pump pumps reagents through the fluidic cartridge. In some embodiments, the pump is actuated by an actuator and wherein upon the actuation the reagents are passed through a filter. In some embodiments, the pump comprises a bellows structure. In some embodiments, the pump comprises a stepper motor. In some embodiments, the pump comprises a blister pouch. In some embodiments, processing the fecal sample comprises extracting and labeling nucleic acids in the fecal-derived microbiome.

[0010] In another aspect, the present disclosure provides an automated system for assaying a biological sample of a subject comprising: (a) a fluidic cartridge comprising (i) a plurality of fluidic processing elements for processing the biological sample of the subject to extract nucleic acid molecules and for labeling the extracted nucleic acid molecules, and (ii) a microarray comprising probes capable of hybridizing with the labeled nucleic acid molecules; and (b) an automation device capable of automating the fluidic cartridge, imaging the microarray and transmitting data from an image of the microarray to a remote computer, wherein the fluidic cartridge is registerable to the automation device.

[0011] In some embodiments, the plurality of fluidic processing elements comprises prefilled reagents. In some embodiments, the biological sample is a fecal sample. In some

embodiments, the probes in the microarray are capable of hybridizing with the labeled nucleic acid molecules derived from fungi, viruses, protozoa, archaea, bacteriophages, plants, human host, animals, or bacteria.

[0012] In some embodiments, the fluidic cartridge comprises a filter through which the biological sample is passed through to extract nucleic acid molecules. In some embodiments, the filter is a monolithic porous plug with a defined pore size. In some embodiments, the defined median pore size of the filter is about 1 nanometer to about 200 micrometer in diameter. In some embodiments, the filter is chemically treated. In some embodiments, the automated system further comprises a plurality of filters. In some embodiments, the automated system further comprises at least 2 filters. In some embodiments, the automated system further comprises at least 3 filters. In some embodiments, the automated system further comprises at least one filter of the plurality of filters is configured to remove enzymatic inhibitors. In some embodiments, the automated system further comprises at least one filter of the plurality of filters is configured to grind the fecal sample. In some

embodiments, the automated system further comprises at least one filter of the plurality of filters is configured to capture nucleic acid molecules for the nucleic acid analysis. In some embodiments, the filter is configured to retain beads. In some embodiments, the beads are coated with a chemical moiety for removing enzymatic inhibitors. In some embodiments, the beads are configured to grind the fecal sample. In some embodiments, the beads are coated with a chemical moiety for capturing nucleic acid molecules. In some embodiments, the automated system further comprises a vibratory motor for homogenizing the biological sample. In some embodiments, the automated system further comprises an ultrasonic transducer for homogenizing the biological sample.

[0013] In some embodiments, the automated system further comprises a pump that pumps the reagents through the fluidic cartridge. In some embodiments, the pump is actuated by an actuator and wherein upon the actuation the reagents begin to enter a filter. In some embodiments, the pump comprises a bellows structure. In some embodiments, the pump comprises a stepper motor. In some embodiments, the pump comprises a blister pouch.

[0014] In some embodiments, the remote computer is capable of processing the image data and generating a report for the subject. In some embodiments, the imaging unit comprises a light source for providing an excitation wavelength to the microarray. In some embodiments, the imaging unit comprises a lens for amplifying an emitted signal from the microarray. In some embodiments, the imaging unit comprises a camera for capturing the image of the microarray. In some embodiments, the automated system further comprises a micro-lens per feature of the microarray. In some embodiments, the automated system further comprises a single lens for the microarray.

[0015] In another aspect, the present disclosure provides a microarray comprising a plurality of probes that selectively hybridizes to genomic sequence or transcripts derived from functional or taxonomic identification genes present in a microbiome.

[0016] In some embodiments, at least a portion of probes selectively hybridize to control nucleic acid molecules. In some embodiments, the functional genes are involved in functional processes selected from the group consisting of biogeochemical cycling of Carbon, Nitrogen, Sulphur, Phosphorus, metals, antibiotic resistance, biodegradation of environmental contaminants, nutrient metabolism, fermentation, short chain fatty acid (SCFA) production, CRISPR bacterial adaptive immunity, cell surface receptors, drug metabolism, virulence factors, signaling molecules and precursors thereof, defense systems, transport systems, lactose metabolism, genome replication, gene transcription, transcript translation, pH reduction, pH increase, bacteriophage adsorption, bacteriophage DNA injection,

bacteriophage replication, bacteriophage host defense circumvention, bacteriophage lysis of bacterial host, bacterial and archaeal transformation, bacterial and archaeal conjugation, and stress responses. In some embodiments, the taxonomic identification genes are selected from the group consisting of 16S rDNA genes, genes encoding components of CRISPR system,

18S rDNA genes, internally transcribed spacers found within rDNA genes, genes encoding components of the ribosome translation machinery, genes encoding components of replication machinery, genes encoding components of the transcription machinery, genes located on plasmids, viral genes, bacteriophage genes, protozoan genes, fungal genes, human genes, and transcripts derived from the genes.

[0017] In another aspect, the present disclosure provides a method for performing nucleic acid analysis of fecal-derived microbiome, the method comprising: (a) applying a fecal sample to a fluidic cartridge comprising a nucleic acid extraction unit and a microarray, the cartridge being compatible with an automation device; (b) automating and imaging the cartridge with the automation device; and (c) obtaining information processed by the processor to determine the fecal-derived microbiome.

[0018] In some embodiments, the fecal-derived microbiome is reported to a user. In some embodiments, a treatment based on the fecal-derived microbiome is reported to a user. In some embodiments, a product recommendation based on the fecal-derived microbiome is reported to a user.

[0019] In another aspect, the present disclosure provides a method for analyzing microbiome- derived data of a subject, the method comprising: (a) using a microarray comprising probes capable of hybridizing with nucleic acid molecules to obtain the microbiome-derived data of the subject as a function of time; (b) providing the microbiome-derived data together with annotations to train a machine learning system to measure changes in microbiomes; and (c) employing the trained machine learning system to infer unmeasured annotations of the subject based on the associations between microbiome-derived data and annotations, and generate a report thereon comprising the associated data for delivery to the subject.

[0020] In another aspect, the present disclosure provides a method of analyzing microbiome- derived data of a subject, the method comprising: (a) using a microarray comprising probes capable of hybridizing with nucleic acid molecules to obtain the microbiome-derived data of the subject; (b) providing the microbiome-derived data together with annotations to train a machine learning algorithm to associate the microbiome-derived data with the annotations; and (c) employing the machine learning algorithm to infer unmeasured annotations of the subject based on the associations between microbiome-derived data and annotations, and generate a report thereon comprising the associated data for delivery to the subject.

[0021] In some embodiments, the annotations comprise medical and non-medical annotations associated with the subject or environment. In some embodiments, the machine learning algorithm is capable of classifying the subject’s microbiome-derived data and associated annotations together with data from other subjects with similar microbiome-derived data. In some embodiments, the machine learning algorithm is capable of virtually connecting subjects with similar microbiome-derived data. In some embodiments, the machine learning algorithm is capable of optimizing product recommendations that can be made for subjects with a given microbiome-derived data. In some embodiments, the machine learning algorithm is capable of optimizing sequences of probes for analyzing microbiomes. INCORPORATION BY REFERENCE

[0022] All publications, patents, and patent applications mentioned in this specification are herein incorporated by reference to the same extent as if each individual publication, patent, or patent application was specifically and individually indicated to be incorporated by reference.

BRIEF DESCRIPTION OF THE DRAWINGS

[0023] The novel features of the disclosure are set forth with particularity in the appended claims. A better understanding of the features and advantages of the present disclosure will be obtained by reference to the following detailed description that sets forth illustrative embodiments, in which the principles of the disclosure are utilized, and the accompanying drawings of which:

[0024] FIG. 1 shows an overview of an automated system comprising device fluidic cartridge, automation device and a data communications interface.

[0025] FIG. 2 shows an exemplary workflow for use of an automated system.

[0026] FIG. 3 shows a schematic of a fluidic cartridge and an automation device in an automated system.

[0027] FIG. 4 shows an exemplary flowchart of processes that may occur in a fluidic cartridge.

[0028] FIG. 5A shows a sample collector and a first pump in an expanded state. FIG. 5B shows a sample collector and a first pump in a contracted state.

[0029] FIG. 6 illustrates dispensing of reagents into a fluidic cartridge using a bellows pump.

[0030] FIG. 7 shows a monolithic porous plug in a microchannel.

[0031] FIG. 8A shows a diversion valve in a top view of a fluidic cartridge. FIG. 8B shows a diversion valve in a cross-sectional view of a fluidic cartridge.

[0032] FIG. 9A shows identification of a microorganism or a group of microorganism using taxonomic identification probes. FIG. 9B shows identification of a gene or a group of genes using functional gene probes.

[0033] FIG. 10 shows a workflow of an infant microbiome profiling.

[0034] FIG. 11 illustrates an example data infrastructure for use with machine learning algorithms that can associate user annotations and databases to recommend products, and/or probe sequences for use in a microarray, and/or update databases and associations in real- time. [0035] FIG. 12A shows a theoretical composition of a microbial community standard. FIG. 12B shows results of DNA extraction by an automated microfluidic system compared to an industry standard PowerSoil kit.

[0036] FIG. 13A shows abundances of microorganisms when assaying using automated DNA extraction and a custom microarray. FIG. 13B shows abundances of microorganisms using commercially extracted DNA.

[0037] FIG. 14 shows a configuration of fluidic cartridge device.

DETAILED DESCRIPTION

[0038] The present disclosure provides devices, systems, and methods for determining the microbiome profile (hereafter referred to as the“microbiome”) of a sample by performing nucleic acid analysis. The present disclosure provides portable devices and systems that can be used for automatic assessment of a microbiome from a variety of samples, such as stool, soil, plants, animals, cell cultures, waste water, treated water, probiotics, providing sample- to-answer in a short time without using traditional benchtop instruments. The present disclosure provides a portable device that can be used for regular monitoring of a microbiome using samples that can be obtained easily with non-invasive methods. For example, a gut microbiome can be regularly monitored using stool samples with the present systems.

Regular monitoring of the gut microbiome can be useful in characterizing changes due to an intervention, such as medication, diet, prebiotics, probiotics, travel, or change in lifestyle, as well as determining appropriate responses to these changes. The present system can comprise a suite of machine learning algorithms for identifying indications of medical conditions, healthy state, pharmacological responses to a therapeutic intervention, indications of non medical conditions treatable by non-prescription means, or responses to non-therapeutic interventions. Such machine learning algorithms can be used for identifying patterns within two or more temporally distinct samples from an individual, for identifying patterns using a collection of samples from a population, for identifying differences between two or more distinct populations, and/or for identifying two or more distinct populations within a collection of samples. Such machine learning algorithms can be used to recommend actions and/or products likely to improve the health of the microbiome.

[0039] In an aspect, the present disclosure provides an automated system for determining the microbiome profile or composition of a sample. An exemplary automated system is illustrated in Fig. 1. An automated system 100 comprises a fluidic cartridge 102 for processing a fecal sample 104 and an automation device 106 for effectuating automated processing of a sample 104 deposited into the fluidic cartridge 102. The fluidic cartridge 102 can include fluidic processing elements, such as fluidic channels, reagents, filters, pumping components for extracting nucleic acid molecules from the fecal sample 104 and at least one microarray for detecting the extracted nucleic acid molecules. The automation device 106 can include actuation hardware for automating sample processing in the fluidic cartridge 102 and imaging hardware for collecting data from the microarray. The automation device can automate the sample processing and capture an image of the microarray using imaging hardware, e.g., a camera, once the fluidic cartridge 102 is registered with the automation device 106. The fluidic cartridge 102 can be registered by placing the fluidic cartridge 102 in a tray 108 for feeding into automation device. The automation device 106 can capture an image of the microarray to collect the data for transmission to a processor (not shown). The processor can process the image for characterization of microorganisms in the microbiome derived from the fecal sample 104. The results from such analysis can be reported to a user, such as by sending to a portable or desktop device 110, for viewing the results. The report can include: information on microbiome; diagnosis, comparison to healthy microbiome and/or with microbiome before any intervention; comparison with microbiomes

representative of a population and/or condition, recommendations on diet, medications, probiotics, enriched formula, nutritional supplements, other products, and/or evaluation of intervention results.

[0040] Fig. 2 shows an exemplary workflow 200 for using an automated system to assay and analyze a microbiome. A user collects a sample and aliquots the sample, if needed 202. The user adds the sample to a fluidic cartridge and inserts the fluidic cartridge into an automation device, insertion of which actuates the automation device for further processing of the sample 204. Raw unprocessed nucleic acid sequence data is collected by the automation device and transmitted to a server 206. The image data is then processed in the server, remote or local, to generate results 208. The results are delivered to the user 210.

[0041] In an aspect, the fluidic cartridge comprises fluidic processing elements and a microarray. A fluidic cartridge can process a sample to extract nucleic acid molecules. The sequences of extracted nucleic acid molecules can be detected by a microarray, coupled to the fluidic cartridge. Fluidic processing elements in a fluidic cartridge can process a variety of samples to determine microbiome of the sample. Non-limiting examples of the samples that a fluidic cartridge can process include stool, soil, urine, tears, sputum, mucus, perspiration, lactation products, swabs from body cavities (e.g., vaginal, oral), plants, food (e.g., fish, fruits, vegetables), wood, environmental samples, waste water, treated water, probiotic pills, drops, enriched foods, etc. In an aspect, a fluidic cartridge can perform nucleic analysis of fecal-derived microbiome, wherein the fluidic cartridge is capable of processing a fecal sample and performing nucleic acid analysis of the fecal-derived microbiome.

[0042] In some cases, a fluidic cartridge can process raw samples (e.g., non-processed) applied directly to the cartridge. For example, a fecal sample can directly be applied to a fluidic cartridge for determining the fecal-derived microbiome, without requiring any prior processing (e.g., homogenizing in buffer solution).

[0043] A sample can be collected for application to a fluidic cartridge or an automation device via a sample collector. A sample collector can be separately supplied with the automated system. A sample collector can be coupled to either a fluidic cartridge or an automation device such that the sample collector can be detached prior to use. A sample collector can include a collection surface and a handle. A collection surface can have a substantially concave or a flat surface for sample collection. For example, a collection surface can have a concave surface. A collection surface can have a volume indicator for collecting a sufficient amount of sample required for a fluidic cartridge. A collection surface can be coated with a chemical moiety for easy dispensing of a sample onto a fluidic cartridge. For example, a collection surface can be coated with chemicals to provide a hydrophobic surface. In another example, a collection surface can be coated with chemicals to provide a non reactive or inert surface. A sample collector can be sterilized and can be wrapped in a sealed bag to keep the sample collector sterile prior to use. In some cases, a sample collector can have a lid for sealing the sample collector. For example, a lid can be a screw-cap or a snap- cap that can provide an airtight seal to the collector. In some cases, a lid may not provide an airtight seal in order to allow release of gases.

[0044] A sample collected by a sample collector can be applied to the microfluidic cartridge for further processing. In some cases, a sample collected by a sample collector can be applied to a fluidic cartridge using an additional tool, such as spoon, toothpick, stick, etc. In some cases, a sample collector can be directly coupled to a fluidic cartridge for applying a sample to the fluidic cartridge. For example, a collection surface of a sample collector can be coupled to a sample inlet opening on the fluidic cartridge. A handle of a sample collector can be detached once a collection surface is in communication with a fluidic cartridge. A sample inlet opening on the fluidic cartridge can be sealed once a sample is applied. For example, a sample inlet opening can be sealed by using a peel or a foil.

[0045] A sample can be automatically processed once applied to a fluidic cartridge, when the fluidic cartridge is inserted into an automation device. Automatic processing in a fluidic cartridge can be triggered by closing of the tray (e.g., 108 in Fig. 1) of an automation device. Automatic processing in a fluidic cartridge can be triggered by manual activation of an automation device, such as by a button, switch, or the like. Automatic processing in a fluidic cartridge can be triggered at a pre-determined time, such as after 5 seconds, 30 seconds, 1 minute, 5 minutes, or more.

[0046] A fluidic cartridge can comprise fluidic processing elements and at least one microarray. A fluidic cartridge can be configured for extracting nucleic acid molecules from a fecal sample. A microarray can be configured to selectively hybridize and detect sequences of the extracted nucleic acid molecules from a fecal sample. A microarray can be fluidically coupled downstream to the fluidic cartridge.

[0047] A fluidic cartridge comprises a plurality of reaction chambers through which a sample automatically passes. Reaction chambers can be supplied with reagents, e.g., by pumping the reagents into the reaction chamber, for carrying out sample processing reactions. Reaction chambers can comprise filters for removing contaminants, grinding the sample, binding to nucleic acid molecules, and/ filtration purposes. Reaction chambers can be connected via micro channels for transferring reagents, sample, or waste products during sample processing. The number of reaction chambers in a fluidic cartridge can be at least 1 to about 10. In some cases, a fluidic cartridge can include at least 2, 3, 4, or 5 reaction chambers. Reaction chambers can be arranged in a series. Reaction chambers can be in parallel arrangement or any other arrangements. Reaction chambers can be of the same size or shape. Reaction chambers can be of different sizes or shapes.

[0048] Reaction chambers in the fluidic cartridge can process a sample to extract nucleic acid molecules from the sample using sample processing reagents. A series of reaction chambers can sequentially process a sample and components thereof by transferring sample through various chambers that are supplied with different reagents. For example, a sample can be collected in an input reaction chamber that can be passed through reaction chambers to remove inhibitors, to extract and purify nucleic acid molecules for labeling the nucleic acid molecules for transfer to a microarray.

[0049] A schematic of a fluidic cartridge and an automation device in an automated system is illustrated in Fig. 3. The fluidic cartridge 300 comprises fluidic processing elements 302 and a microarray 304. The fluidic cartridge 300 comprises a plurality of reaction chambers for processing a sample 306 to extract nucleic acid molecules. The sample 306 can be collected by a sample collector by a user. The sample collector can include a collection surface 308 and a handle 310. The collection surface 308 can be coupled to a sample inlet opening 312 in the fluidic cartridge 300. The collection surface 308 can provide a seal between the sample collector and the sample inlet opening 312. The handle 312 of the sample collector can be detached after coupling the sample collector to the sample inlet opening 312. The sample inlet opening 312 opens into a sample input reaction chamber 314. The sample input reaction chamber 314 can be supplied with sample dissolution reagent 316 that can automatically be pumped into the sample input reaction chamber 314. The dissolution reagent 316 can be mixed with the sample 306 either passively, e.g., by gravity or actively, e.g., by activity of a homogenizer 318. The input reaction chamber 314 can transform (e.g., by mixing sample with the dissolution reagent) the sample 306 into a slurry that can be transported to a sample cleaning chamber 320, via a first microchannel 322, for removing enzymatic inhibitors (e.g., PCR inhibitors). The cleaned sample can be transported to a nucleic acid extraction chamber 324, via a second microchannel 326, for extracting nucleic acid molecules. The sample extraction chamber 324 can be supplied with nucleic acid extraction reagents 328, such as lysis reagent. Extraction of nucleic acid molecules in the sample extraction chamber 324 can be facilitated by applying physical forces, such as vibratory motion using a homogenizer 318, and/or passing the sample 306 through beads. The extracted nucleic acid molecules are transferred to a nucleic acid binding chamber 330, wherein the nucleic acid molecules can be bound to a filter or a matrix, via a third microchannel 332. The bound nucleic acid molecules can be washed with wash buffer and eluted with elution buffer 334, for transferring to a labeling reaction chamber 336 via a fourth microchannel 338. The labeling reaction chamber 336 can be supplied with labeling reagents 340 for labeling the nucleic acid molecules. The labeling reagents 340 can comprise“spike-in” control nucleic acid molecules with known sequence(s) and known concentration(s) for subsequent quality control, normalization, or quantification (e.g., absolute or relative). The labeling reaction chamber 336 can be supplied with heat or chemical denaturants for denaturing nucleic acid molecules. The reaction chambers can be connected to a waste chamber 342 for collecting waste products from the reaction chambers via micro channels. The labelled nucleic acid molecules can be transferred to the microarray 304, via a fifth microchannel 344 for hybridizing with probes.

Hybridization of nucleic acid molecules with a probe can be indicative of presence of a specific microorganism in the sample. The hybridization signal can be detected using an automation device (shown by dotted lines) of the automated system 346 by registering the fluidic cartridge 300 to the automation device. The automation device can include a light source 348 for providing an excitation light, a camera 350 for capturing an emitted signal in an image, and a transmitter 352 for transferring the image to a server for further processing. The sample processing in various reaction chambers of the fluidic cartridge is shown in a flowchart in Fig. 4.

[0050] A fluidic cartridge can process a sample sequentially to extract and label nucleic acid molecules for further analysis by a microarray. The sequential processing of a sample can be achieved by using actuators and pumps for sequentially releasing reagents to appropriate reaction chambers upon actuation of fluidic cartridge by an automation device. In some cases, filters can be planar mesh structures (e.g., filters by Millipore, Germany), monolithic inserts (e.g., filters by Porex™, Georgia or 3M Empore™ by Sigma-Aldrich). Actuators for pumps can be fabricated from stepper motors or solenoids available from many suppliers of small mechanical components, e.g. Portescap. Pumps, valves, and sample collectors can be commercially available and/or bespoke designs for low-cost integration with the fluidic cartridge and automation device. The sequential processing can be initiated by the insertion of a fluidic cartridge into the automation device, by a startup operation by the user, e.g. a switch, push-button, command from an software application in a smartphone, or any other common starting operation either physically on the automation device or communicated electronically from any common device. The sequence will be determined by a program in the automation device, which can be permanently installed, downloaded via electronic communication, requested electronically by the automation device using the code on a fluidic cartridge, or any combination of the above.

[0051] As shown in FIG. 5A in an exemplary embodiment of a fluidic cartridge registered with an automation device (not shown), the fluidic cartridge 500 can be coupled to a plurality of bellows pumps that can be actuated by an actuator (e.g., stepper motor) upon application of the sample. The plurality of bellows pumps can be permanently or detachably coupled to the fluidic cartridge. The fluidic cartridge 500 comprises a plurality of bellows pumps that are coupled to a plurality of plunger pumping elements 502. A combination of a bellows pump and a plunger pump can be associated with a single reaction chamber in order to provide reagents and fluidic drive required in the reaction chamber. Sample 504 can be collected in a sample collector 506 and applied to the fluidic cartridge 500. Upon application of the sample 504, the fluidic cartridge 500 can be registered (e.g., by inserting into an automation device) with an automation device. The sample collector 504 can be coupled to the fluidic cartridge 500 and the sample 504 can be released into the fluidic cartridge 500 by pushing the collector 506 to dispense the sample 504, as shown in FIG. 5A. As shown in FIG. 5B, the sample 504 is pushed into a first reaction chamber 508. Once the sample is in the first reaction chamber 508, a first pump 510 can be actuated, releasing dissolution/lysis reagent. Dissolution/lysis reagent dissolves the sample to form a slurry and lyses cells in order to extract nucleic acid molecules from the slurry. Then, a second pump 512 can be actuated to release wash buffer in order to wash away impurities. Next, a third pump 514 can be actuated to release elution buffer to elute nucleic acid molecules for transferring into a labeling chamber 516. Next, a fourth pump 518 can release labeling reagents (and optionally,“spike-in” control

oligonucleotides) for labeling the eluted nucleic acid molecules. The labeled nucleic acid molecules can be transferred to a microarray 520 for hybridizing with probes.

[0052] Reagents required for a fluidic cartridge can be prefilled in the cartridge for supplying to the reaction chambers in the cartridge. In some cases, reagents can be filled into blister pouches. Blister pouches can be coupled to a fluidic cartridge. Blister pouches can be pierced to release the reagents upon actuation of the fluidic cartridge. Once pierced, reagents can be in a fluidic communication with the appropriate reaction chamber via a fluidic channel. For example, upon piercing of a blister pouch containing dissolution/lysis reagent, the dissolution/lysis reagent can flow through a fluidic channel to an input reaction chamber to provide reagents for dissolving a sample. In some cases, reagents can be filled into a bellows pump. A bellows pump can be coupled (e.g., permanently or detachably) to a fluidic cartridge. A bellows pump can be contracted to release the reagents upon actuation of the fluidic cartridge. A bellows pump can be sealed at a nozzle end from which fluid can be dispensed. A bellows pump can be sealed by a pierceable material, such as a foil or peel, at the nozzle end. A seal at the nozzle end of a bellows pump can be broken upon actuation of the fluidic cartridge. In some cases, reagents can be stored in liquid form. In some cases, reagents can remain in a powder form, such as lyophilized reagents, until they come in contact with a sample or liquid reagent.

[0053] Flow in micro channels in a fluidic cartridge can be controlled by micro pumps for automatic processing of a sample. For example, micro pumps can be used for pumping reagents to reaction chambers, pumping sample between the chambers during sample processing, or pumping waste products to a waste chamber. In some cases, flow in micro channels can be controlled by pressure-driven flow mechanism provided by a syringe pump, capable of achieving uniform flow with less pulsation. In some cases, flow in micro channels can be controlled by self-priming micro pumps having efficient power consumption. In some cases, flow in micro channels can be controlled by low flow peristaltic pumps that can provide variable flow. In some cases, flow in micro channels can be controlled by a positive displacement pump. For example, flow micro channels can be controlled by a bellows pump that can pump the fluid through the micro channels by contracting the bellows. A bellows pump can create a steady or pulsating flow. A bellows pump can pump a volume of fluids through micro channels. A bellows pump can be operatively coupled to a plunger pump or pumping elements. A bellows pump can be permanently coupled to a plunger pump or pumping elements. A bellows pump can be detachably coupled to a plunger pump or pumping elements. In some other cases, flow in micro channels can be controlled by a plunger pump or pumping elements. In another example, flow in micro channels can be controlled by low flow piston pump systems that can deliver fluids smoothly and precisely. In some cases, micro pumps can be coupled with a metering pump mechanism for moving a precise volume of fluid in a specified time period providing an accurate volumetric flow rate.

[0054] As shown in FIG. 6, reagents can be stored in a blister pouch 602 housed inside a bellows pump 604 coupled to a fluidic cartridge 606. The blister pouch 602 can be pierced at a nozzle end 608 to release the reagents into a porous plug 610 by compressing the bellows pump 604. Once the reagents pass through the porous plug 610, the fluid can move towards the right, as shown by the arrows.

[0055] A fluidic cartridge comprises filters through which a sample can be passed to extract nucleic acid molecules. A filter can be placed inside one or more reaction chambers. A filter can be configured depending on the reaction chamber. For example, a filter, such as a planar filter, can be included between an input reaction chamber and a contaminant removal chamber to remove undissolved sample materials. A filter can be included to block the passage of beads to subsequent chambers; these beads can be utilized as grinding and lysing elements in the chamber and prevented from flowing downstream. Beads can also be fabricated of or coated with materials that provide a functional advantage, e.g. to capture contaminants and limit their passage to subsequent reactions; alternatively or in addition to, coated or fabricated beads can be used as a reaction element, e.g. to capture and subsequently release nucleic acids, even as the beads are held immobile by the filter. Beads can also be captured by a coarse filter to assemble a finer structure bulk filter. A filter in a contaminant removal chamber can be treated with chemicals for removing enzymatic inhibitors, such as PCR inhibitors. A filter can be treated with chemicals for lysing cells, such as bacterial cells for example. A filter in a nucleic acid binding reaction chamber can be configured to bind to nucleic acid molecules and elute with the application of elution buffer. In some cases, chemicals can be incorporated into a porous filter. In some cases, a chemical moiety can be linked to the beads comprised in a filter. Chemicals or chemical moieties can be selected based on the composition of a sample. For example, a fecal sample can contain enzymatic inhibitors, such as complex polysaccharides, bile salts, lipids, urate, humic acid, etc. Based on the inhibitors, a filter can be treated with polyvinylpyrrolidone (PVP), bovine serum albumin (BSA), cetrimonium bromide, T4 gene 32 protein (gp32), or similar chemicals to remove enzymatic inhibitors from the fecal sample. In some cases, chemicals or chemical moieties can be selected based on the reaction carried out in a reaction chamber. For example, when a sample is in an input reaction chamber, then the sample can be passed through a filter capable of removing enzymatic inhibitors prior to homogenization. Such a filter can comprise PVP, BSA, or the like for removing the inhibitors. Upon homogenization, a sample can be passed through a nucleic acid extraction chamber for lysing the cells, such as bacterial, archaeal, and/or eukaryotic cells, and extracting nucleic acid molecules. A filter in a nucleic acid extraction chamber can comprise lysis reagents, such as hemolysin, mutalysin, detergents or the like. A filter in a nucleic acid extraction chamber can comprise nucleic acid binding chemicals for binding to the extracted nucleic acid molecules. Non-limiting examples of such binding chemicals include silica, cellulose, or ion exchange (e.g., salt). Extracted nucleic acid molecules can be washed to remove any contaminants. Extracted nucleic acid molecules can be eluted for releasing into a labeling chamber for labeling the extracted molecules.

[0056] In some cases, a filter can be a monolithic porous plug with a defined pore size. In some cases, pores of a filter can be uniform in size. In some cases, pores of a filter can be of different sizes. Pore size of a filter can range from about 1 micrometer (pm) to about 200 pm. In some cases, pore size of a filter is about 0.5 pm to about 200 pm. In some cases, pore size of a filter is at least about 5 pm to about 150 pm, at least about 10 pm to about 100 pm, at least about 20 pm to about 50 pm or at least about 30 pm to about 40 pm. In some cases, a filter can comprise beads, such as garnet beads, retained by a mesh. Non-limiting examples of beads in a filter include garnet, silica, or zirconium beads. Size of the beads in a filter can range from about 1 pm to about 1000 pm). In some cases, size of the beads in a filter can be uniform prior to applying vibratory motion during homogenization of the sample. After vibratory motion, the beads in a filter can have different sizes. For example, the beads can have relative sizes that can be grouped into: large, medium, and small. In some cases, a filter can comprise a combination of beads retained by a mesh and a monolithic porous plug.

[0057] Filters can include commonly used materials. In some cases, filters can be planar mesh structures (e.g., filters by Millipore, Germany), monolithic inserts (e.g., filters by Porex™, Georgia or 3M Empore™ by Sigma-Aldrich). Filters for removing PCR inhibitors, cross-linked polyvinylpyrrolidone matrix, can be purchased from Sigma-Aldrich, Fisher Scientific, or from Claremont Biosciences. Sample lysis beads can be garnet beads 1.0 mm (Biospec Products), 0.5mm (OPS Diagnostics) or 0.7mm (OMNI International). Sample lysis beads can be silica or zirconium beads in a mixture of sized (OPS Diagnostics). Many suppliers of silica beads can be suitable as DNA bind-and-wash beads, including beads by Thomas Scientific and Sigma-Aldrich. Ceramic beads for cleaning extracted nucleic acids can be purchased from Claremont Biosciences.

[0058] As shown in FIG. 7, a fluidic cartridge 700 can have a porous plug filter 702. The porous plug filter 702 can have pores of varying sizes and shapes. Fluid can be passed through the porous plug 702 to extract nucleic acid molecules. Flow of the fluid is shown by the arrows.

[0059] Sample processing in a fluidic cartridge can be improved by using a homogenizer. A homogenizer can facilitate mixing of the sample with dissolution buffer to form a slurry. A homogenizer can facilitate grinding of a sample for extraction of nucleic acid molecules by lysing cells, such as bacterial cells. A homogenizer can be coupled to a fluidic cartridge, such as near the fluidic processing elements, sample inlet reaction chamber, nucleic acid extraction reaction chamber, or on top or bottom of a fluidic cartridge. A homogenizer can be permanently or detachably coupled to a fluidic cartridge. A homogenizer may not be attached to the automated device, but can be separate from a fluidic cartridge, vortexer, for example.

[0060] Homogenizers can be based on bead mills, sonication, mechanical, high pressure, optical, or other physical forces. In some cases, a homogenizer can utilize a vibratory motor that can be coupled to a fluidic cartridge for providing vibratory motion. In some cases, a homogenizer can be an ultrasonic transducer that provides sonication energy. In some cases, a homogenizer can be a vortexer that can homogenize a sample by vortexing a fluidic cartridge. In some cases, a homogenizer can include an impeller for blending a sample in a microfluidic cartridge. Homogenization can be accompanied by two or more mechanisms.

For example, a vibratory motor can be used in conjunction with bead mills for grinding a sample. Similarly, an ultrasonic transducer can be used with bead mills. Homogenization with two or more mechanisms can be useful for homogenizing fibrous samples, stool sample, for example.

[0061] Controlling flow in micro channels of a fluidic cartridge via micro valves can be useful for automatic processing of a sample. For example, flow can be controlled to direct the flow to an appropriate reaction chamber, to direct the flow of waste products to a waste chamber, to obstruct the reverse flow between the reaction chambers, or to block the reverse flow from a waste chamber to reaction chambers. Micro valves can be active micro valves that can adjust the flow as a result of elastomeric deflections. In some cases, active micro valves can be based on hydraulic deflection that in turn is based on magnetic actuation by solenoids. For example, an active micro valve can be a diversion valve that is based on hydraulic deflection actuated by a solenoid. In some other cases, an active micro valve can be based on torque-actuated deflection using small machine screws. In some cases, passive approaches, which leverage differences in fluid behavior from varying channel geometries in capillary systems, can be employed. The passive approaches do not require external power and additional parts. For example, delay valves can merge smaller channels into larger channels to allow smooth collection of incoming fluid at different flow rates. On the contrary, stop valves can reduce the width of a channel using a restriction and enlarge it abruptly to reduce capillary pressure of fluid to zero. As shown in FIG. 8A, an active micro valve can be used in a microfluidic channel to control flow volume and flow direction to one or more output channels. A diversion valve 800 can be used to direct flow of fluid from input channel 802 to output channel 2 804, instead of output channel 1 806. As shown in Fig. 8B, Diversion valve 800 can obstruct the flow of the fluid to output channel 1 806, upon actuation.

Diversion valve 800 can be actuated by actuator, a solenoid or a stepper motor, for example. In this case, flow of the fluid can be accelerated in input channel 802 and directed as a jet toward the output channels. The jet can be deflected into output channel 2 804 due to diversion valve 800.

[0062] A fluidic cartridge can be operatively coupled to an actuator for incorporating moving or controlling mechanisms into the fluidic cartridge. An actuator can be coupled to an automation device. In some cases, an actuator can be used for controlling micro pumps, micro valves, homogenizer, etc. For example, an actuator can be used for controlling a homogenizer such that the homogenizer can homogenize a sample upon actuation. Non limiting examples of actuators include pneumatic, hydraulic, mechanical, thermal, magnetic, piezoelectric, stepper motor, spring or electric.

[0063] Sample pH can be a useful indicator of health of an individual, ecosystem, or the like. For example, fecal pH of a breastfed infant can be lower than formula-fed infants due to a microbiome rich in Bifidobacterium. The automated system disclosed herein can process raw, non-processed samples and can be equipped to measure pH of a sample. For example, the automated system can be equipped with a pH indicator for measuring pH of a sample, fecal sample, for example. The automated system, preferably before a sample is processed, can be equipped with a pH indicator. For example, a sample collector can be treated with a color changing pH indicator, such as phenol red that change color in response to pH of a sample or aliquot thereof. In some cases, sample dissolution reagent in a fluidic cartridge can comprise a color-changing pH indicator. pH of a sample can be determined by a user by recording the color of the sample before and after the sample comes in contact with the color-changing pH indicator. A change in color can be recorded by taking a picture and/or by noting the color of a sample. A change in color can be compared with a standard color chart with a color range indicative of pH values for a given indicator. A change in color can be digitally recorded by a camera in an automation device and processed by a processor to determine pH of a sample.

[0064] The automated system can measure the concentration of extracted nucleic acid molecules. Such quantification can be useful for assessing quality and quantity of extracted nucleic acid molecules. For example, an aliquot of extracted nucleic acids can be transferred microfluidically to a separate chamber containing nucleic acid-staining dyes (e.g. SYBR green), whereupon an image can be captured, degree of staining can be quantified, and extrapolated for the remaining sample. For example, the extracted nucleic acid can be quantified by measuring UV absorbance at various wavelengths, such as 260 nm which is indicative of both DNA and RNA concentrations.

[0065] A fluidic cartridge can label extracted nucleic acid molecules and optionally include “spike-in” control oligonucleotides, if present, in a labelling reaction chamber prior to transferring to a microarray for detection of sequences of nucleic acid molecules. Extracted nucleic acid molecules may not be amplified using polymerase chain reaction. Extracted nucleic acid molecules may be amplified using polymerase chain reaction. Preferably, extracted nucleic acid molecules are not amplified prior to labeling the extracted nucleic acid molecules. Extracted nucleic acid molecules can be fluorophore-, silver-, electrochemical-, or chemiluminescence-labeled. In some cases, extracted nucleic acid molecules can be labelled with fluorophores, such as Cy3 or Cy5. Extracted nucleic acid molecules can be labeled by primer extension, incorporating labeled nucleotides from random or specific primers using DNA polymerase. Extracted nucleic acid molecules can be labeled by a nick translation procedure by first nicking nucleic acid strand(s) prior to extending with labeled nucleotides. Labeled nucleic acid molecules can be generated via reverse transcription such that a label can be added either during transcription or amplification. Labeled nucleic acid molecules can be denatured prior to hybridization with probes on a microarray. Denaturation of labeled nucleic acid molecules can be carried out by using chemical and/or physical methods. Non limiting examples of chemical denaturants include acetic acid, HC1, nitric acid, NaOH, DMSO, formamide, guanidine, sodium salicylate, propylene glycol, and urea. Non-limiting examples of physical denaturants include heat, beads, sonication, and radiation. A fluidic cartridge can be equipped to carry out denaturation. For example, a labelling reaction chamber can be supplied with chemical denaturants, if chemical denaturation is being carried out. A labelling reaction chamber can be heated, if physical denaturation is being carried out. In some cases, a fluidic cartridge can be heated using an external heater. In some cases, a heater can be integrated with a fluidic cartridge. In some cases, a heater can be permanently or detachably coupled to a fluidic cartridge

[0066] A fluidic cartridge can include at least one microarray capable of performing a nucleic acid analysis reaction. In some instances, probes are capable of hybridizing with labeled nucleic acid molecules to perform nucleic acid analysis in order to determine microbiome. In some cases, a microarray can be coupled to a microfluidic cartridge. A microarray can be integrated into a microfluidic cartridge. A microarray can be separate from a fluidic cartridge. In this case, the labeled nucleic acid molecules from the fluidic cartridge can be transferred manually to the microarray. In some cases, the labeled nucleic acid molecules can be automatically transferred to a microarray from a fluidic cartridge.

[0067] In some cases, a microarray comprises a plurality of probes that selectively hybridizes to genes or transcripts present in a microbiome. Such functional genes are involved in processes such as, e.g., biogeochemical cycling of Carbon, Nitrogen, Sulphur, Phosphorus, metals, antibiotic resistance, biodegradation of environmental contaminants, nutrient metabolism, fermentation, short chain fatty acid (SCFA) production, CRISPR bacterial adaptive immunity, cell surface receptors, drug metabolism, virulence factors, signaling molecules and precursors thereof, defense systems, transport systems, lactose metabolism, genome replication, gene transcription, transcript translation, pH reduction, pH increase, bacteriophage adsorption, bacteriophage DNA injection, bacteriophage replication, bacteriophage host defense circumvention, bacteriophage lysis of bacterial host, bacterial and archaeal transformation, bacterial and archaeal conjugation, and/or stress responses. For example, a microarray can comprise at least 10, 20, 50, 100, 200, 500, 1000, 2000, 5000, or 10,000 different probes that selectively hybridize to functional genes.

[0068] In some cases, a microarray can include control probes that selectively hybridize to positive, negative, normalization, and/or quantification nucleic acid molecules. Such molecules can be naturally occurring molecules (e.g. negative controls corresponding to transcripts, genes and/or genomic sequences unique to organisms highly unlikely to be found in microbiomes, such as deep sea or jungle animals), and/or exogenously introduced synthetic molecules (e.g. positive controls corresponding to“spike-in” oligonucleotides of known sequence(s) and defined concentrations, negative controls corresponding to sequences not found in any known genomes or metagenomes). For example, a microarray can comprise at least 10, 20, 50, 100, 200, 500, 1000, 2000, 5000, or 10,000 different probes that selectively hybridize to control sequences.

[0069] Alternatively, or in addition to functional genes, a microarray can comprise a plurality of probes that selectively hybridize to taxonomic identification genes. Non-limiting examples of taxonomic identification genes include 16S rRNA genes or rDNA, genes encoding components of CRISPR system, 18S rRNA genes or rDNA and/or derived transcripts, internally transcribed spacers found within rRNA genes or rDNA, components of ribosome translation machinery, components of transcription machinery, components of replication machinery, genes located on plasmids, viral genes, bacteriophage genes, protozoan genes, fungal genes, human genes, and transcripts of human genes. In some instances, a microarray can comprise at least 10, 20, 50, 100, 200, 500, 1000, 2000, 5000, or 10,000 different probes that selectively hybridize to taxonomic identification genes.

[0070] In some cases, a single probe can be used to identify a microorganism. For example, a taxonomic identification gene, such as 16S rRNA gene or rDNA, or a functional gene, such as antibiotic resistance gene, can be used to identify a given microorganism or group of microorganisms (e.g. at the level of strain, species, genus, family, order, class, phylum, kingdom). In some cases, more than one probe can be used to identify a microorganism for improving sensitivity of detection. Probes for identifying a given microorganism can be selected from taxonomic identification genes, functional genes, or a combination thereof.

[0071] As illustrated in FIG. 9A, taxonomic identification probes can be used for identification of a given microorganism or a group of microorganisms at different hierarchical levels, such as domain, kingdom, phylum, class, order, family, genus, species, and/or strain. As illustrated in FIG. 9B, functional gene probes can be used for identification of a given microorganism, group of microorganisms, or enzymatic functions based on different functional categories, such as the three categories shown in the figure: antibiotic resistance, SCFA production, signaling molecules and precursors. The actual list of functional categories can include more functional categories and can be updated over time. Each functional category can comprise sub-categories and sub-categories can be identified based on different functional genes under the category. For example, several sub-categories of the antibiotic resistance category can include genes for kanamycin resistance, ampicillin resistance, etc.

[0072] Sequence of the probes can be shorter (25-mer) or longer (60-mer). For example, a sequence of a probe can be 40-mer. In some cases, all probes on an array can be of the same length. In some cases, probes can have different sequence lengths. In some cases, a single copy of probe is spotted onto a microarray. Multiple copies of probes can be spotted onto a microarray. In some cases, a microarray can include at least about 50 to at most about 10,000 unique probes. For example, a microarray can include 500 unique probes or spots. In other examples, a microarray can include 1000 unique probes or spots.

[0073] In another aspect, the present disclosure provides an automated system for assaying a biological sample of a subject. The automated system can comprise a fluidic cartridge, prefilled reagents for processing the biological sample of the subject to extract nucleic acid molecules and for labeling said extracted nucleic acid molecules, and a microarray comprising probes capable of hybridizing with said labeled nucleic acid molecules; and an automation device capable of automating the fluidic cartridge, imaging the microarray and transmitting data from an image of the microarray to a remote computer, wherein the fluidic cartridge is registerable to the automation device.

[0074] A fluidic cartridge can be inserted into an automation device and the automation device can trigger processing of the sample in the fluidic cartridge and can image a microarray once the device is registered with the imaging unit. An automation device can transmit data from an image of a microarray to a remote computer. An automation device can be activated by any of the methods described elsewhere herein.

[0075] An imager in the automation device can detect and image the intensity of fluorescent spots on a microarray by exciting fluorophores with a light source. Light source can provide an excitation wavelength, such by using an LED light. Light source can have a single or multiple wavelengths. Preferably, light source provides a single wavelength. A light source emitting light of multiple wavelengths may be combined with a filter that narrows the emitted light to a given single wavelength or to a smaller band of wavelengths. Upon excitation, the emitted light from a microarray can be amplified by a lens. In some cases, lens can be a microlens covering a single feature of a microarray. In some cases, lens can be a single lens covering an entire microarray or all features of a microarray. In some cases, lens can comprise multiple lenses formed in a one-dimensional or two-dimensional array. Amplified fluorescence output or emitted signal can be detected and recorded by a camera. In some cases, an imager can be a camera, which can include an image sensor, such as a digital or analog sensor. In an example, a camera includes an analog sensor, such as CMOS or CCD.

An image of a microarray can be transmitted to a processor of a remote computer by a transmitter coupled to an automation device. In some cases, chemiluminescence can be detected using only the imaging hardware of the automation device by exciting the emitting labels affixed to the nucleic acids with the correct chemical substrate. In some cases, electrodes included in the fluidic cartridge can electrochemically interrogate electrochemical labels affixed to the analyte nucleic acids. All of these spatially distinct signals can be processed as image intensity values according to the discussion below.

[0076] Image of a microarray can be processed to analyze each feature of the microarray to determine fluorescence intensities of all the features. Relative quantitation of the fluorescent intensity of a given feature can be compared to the intensity of the same feature under a different condition. Normalization of microarray images can be accomplished by

transforming the matrix of fluorescent intensities using a standard normalization procedure (e.g. quantile normalization, loess normalization, variance stabilization normalization).

Absolute quantification can be accomplished by subdividing the range of signal intensities into bins, each including the signal from one or more positive control probes corresponding to a“spike-in” oligonucleotide positive control, and scaling (multiplying) the intensity values in each bin such that positive control intensities match the defined concentrations of spike-in controls. Identity of a given feature can be determined by its position on a microarray. Based on the identity of a feature, a microorganism (species or strain) or clades of microorganisms (genera, families, orders, classes, phyla, or kingdom) or functional gene in a sample can be taxonomically identified. Preferably, identities of abundant microorganisms in a sample are determined.

[0077] Data from a microarray can be analyzed using standard procedures, for example, as described in US8232055B2. Briefly, signal intensities from specifically and stringently hybridized probe features can be determined by image analyses. Using appropriate background subtraction methods specific intensities can be derived. Statistical analyses, such as determining signal mean, distribution, standard deviation, can be performed. Control features can be used to normalize signal distribution across samples. Either single color hybridization (one sample per array) or two-color hybridization (two samples, frequently a test sample and a control sample) can be performed per array. If two-color hybridization is performed, log ratios of fluorescence intensity (frequently Cy3/Cy5) may be used in lieu of one-color signal intensity to determine abundance of the target sequence.

[0078] Normalized microarray data can be processed bioinformatically to obtain a suite of statistical variables. Probes can be used for taxonomic identification (domains, kingdoms, phyla, classes, orders, families, genera, species, strains) and quantification. Probes can be used to quantify abundance of functional genes belonging to categories and/or pathways (e.g. antibiotic resistance genes, virulence genes, SCFA-production genes, etc.), said genes being unique to a microbial species/strain or shared among higher-level taxa (genus, family, order, class, phylum, kingdom). Statistical variables can include:

1. Taxonomic richness for individual clades at multiple phylogenetic levels (e.g.

number of species present in the sample, number of SCFA-production genes present, etc.).

2. Variance for individual clades at multiple phylogenetic levels (e.g. variance of

summed normalized signal intensity of all species)

3. % change of individual clades at multiple phylogenetic levels for time-series data (e.g. average % change between two samples for all probes unique to a given species)

4. Sums for individual clades at multiple phylogenetic levels (e.g. sum of species-level probe signal intensity for each represented genus)

5. Delta from reference sample (e.g. healthy control)

6. % of reference sample (e.g. healthy control)

7. Alpha diversity for individual clades at multiple phylogenetic levels (Shannon

index: H' = — å ₌₁ (Pi ln( i)), Simpson’s index: D = å ₌₁ p ). Described here: (https://en.wikipedia.org/wiki/Diversity_index)

8. UniFrac diversity for individual clades at multiple phylogenetic levels (both

weighted and unweighted UniFrac, calculated as described here:

(http ://aem. asm . org / content/73/ 5/ 1576. full . pdf+html)

9. Beta diversity between two samples (calculated using Bray-Curtis dissimilarity

S +S —2 C

index: ( BC)i _j =—— Si+ -Sj - [C _Lj = total number of probes in common between the two samples (i.e. determined present in both samples, and Si = total number of probes determined present Sample i] ). Described here:

https://en.wikipedia.org/wiki/Bray%E2%80%93Curtis_dissimilar ity

10. Overlap between samples (i.e. probes determined present that are shared between two or more samples).

11. Overlap between groups (i.e. probes determined present shared among at least X% of samples in group, shared between two or more groups of samples).

12. Stable probes (i.e. probes with a variance below a predetermined threshold across a set of two or more samples), and Variable probes (probes with a variance above said threshold across a set of two or more samples)

13. Signature scores (i.e. the degree to which the intensity signals of relevant probes on the array match published microbiome signatures (e.g. 16S amplicon sequencing, 18S amplicon sequencing, ITS amplicon sequencing, shotgun metagenomics, shotgun metatranscriptomics, etc.) predictive of or associated with asthma

(doi: l0.l038/nm.4l76), childhood atopy (doi: l0. l038/nm.4l76), vaginal or C-section delivery (doi: l0.H0l/gr.233940. H7, doi: 10.3389/fped.2017.00200), type 1 diabetes (http://dx.doi.org/l0.l0l6/j.chom.20l5.0l.00l, doi: l0.l038/ismej.20l0.92), exposure to antibiotics (doi: 10.3389/fped.2017.00200, DOI:

10.1038/NMICROBIOL.2016.24), lack of breastfeeding or formula feeding

(doi: 10. lOOl/jamapediatrics.2017.0378, doi: l0.H86/gb-20l2-l3-4-l52, doi:

10.3389/fped.2017.00200), exposure to household furry pets (DOI 10.1186/s40l68- 0l7-0254-x), inflammatory bowel disease (doi: l0.l038/ctg.20l7.58), celiac disease (https://doi.orgl0.H86/s40l68-0l8-04l5-6), obesity (doi: l0.3390/nu7042237), necrotizing enterocolitis (DOI l0. H86/s40l68-0l7-0248-8), such signatures indicating presence/absence/quantity of pathogenic or beneficial microorganisms, higher-order taxa, genes, transcripts, SNPs, ratios of taxonomic clades (e.g. ratio of total signal derived from Firmicutes phylum and total signal derived from

Bacteroidetes phylum) that contribute to or associate with said indications.).

[0079] Interrogated samples can be associated with user-provided annotations or orthogonal data. Annotations or orthogonal data can be any user-associated data other than the microbiome-derived data. Annotations can include age, ethnicity, diet, chronic conditions, acute conditions, disease, non-medical condition (e.g. colic, diarrhea, eczema, etc.), microbiome disruptors (e.g. C-section birth, antibiotic use by end-user, antibiotic use by mother while in womb, lack of breastfeeding), life events and transitions (e.g. flying to another geographic location, starting daycare, starting solid foods, beginning to crawl, beginning to walk, moving homes, getting a dog, etc.), intervention (when applicable, such as diet, drug intake, vitamin/supplement intake, etc.), time since meal, meal contents, amount of sleep, quality of sleep, time outdoors, amount of physical activity, type of physical activity, stress level, anxiety level, mood, pain level, allergy severity, physical wellbeing, mental wellbeing, etc.

[0080] Interrogated samples can be associated with automatic annotations obtained via the internet, such as geographic location, time of day, weather, altitude, humidity, pollution indices, etc. [0081] Interrogated samples can be associated with automatic annotations obtained via wearables and connected devices, e.g. Fitbit, Apple Watch, other smart watches, heart rate monitors, step counters, breathing monitors, smart cribs (e.g. Snoo), etc.

[0082] Interrogated samples can be associated with automatic annotations obtained via smartphone apps, such as activity logs, feeding logs, diet logs, calendars, etc.

[0083] Interrogated samples can be associated with annotations obtained via genomics products, including SNP genotypes, exome sequencing, amplicon sequencing (individual genes and/or gene panels), whole genome sequencing, chromatin immunoprecipitation- sequencing (e.g. of transcription factors, histone modifications, etc.), bisulfite sequencing of DNA methylation, genome scaffolding, copy number variation, structural variants, liquid biopsy, etc., whether obtained via consumer products (e.g. 23andMe, Color Genomics, etc.) or clinical/medical products (e.g. FoundationOne, Natera Panorama, MSK-Impact, bespoke offerings in hospitals, etc.).

[0084] Interrogated samples can be associated with annotations obtained via

hospital/medical/clinical testing, whether input manually by the user or a health professional, or input automatically via communication with electronic health records, such as heart rate, blood oxygen, blood work (e.g. HDL, LDL, Ale, etc.), EKGs, MRIs, CT scans, X-rays, etc.

[0085] In one instance, an individual fecal sample is analyzed on a server and a report comprising data about the microbiome from that sample is transmitted to an individual (e.g., end user, family of end user, physician of end user, governmental body, wellness

professional, coach, or nutritionist). A report can include information such as: presence, absence and/or abundance of microorganism species, higher-order bacterial taxa, functional genes, clinically relevant associations (e.g., asthma, obesity, type I diabetes, type II diabetes, autism, ulcerative colitis, Crohn’s disease, irritable bowel syndrome, autoimmune disease, leaky gut, food allergies, non-food allergies, eczema, diarrhea, antibiotic-induced diarrhea, atopy, lactose intolerance, high gut pH, colic) with the microorganism profile associated with the sample, correspondence to published microbiome signatures (i.e.

presence/absence/quantity of microorganisms, higher-order microbial taxa, genes, transcripts, etc.) predictive of or associated with asthma (doi: l0.l038/nm.4l76), childhood atopy

(doi: l0.l038/nm.4l76), vaginal or C-section delivery (doi: l0. H0l/gr.233940.H7, doi:

10.3389/fped.2017.00200), type 1 diabetes (http://dx.doi.org/l0.l0l6/j.chom.20l5.0l.00l, doi: l0.l038/ismej.20l0.92), exposure to antibiotics (doi: 10.3389/fped.2017.00200, DOI: 10.1038/NMICROBIOL.2016.24), lack of breastfeeding or formula feeding

(doi: 10. lOOl/jamapediatrics.2017.0378, doi: l0.H86/gb-20l2-l3-4-l52, doi: 10.3389/fped.2017.00200), exposure to household furry pets (DOI 10.1186/s40l68-0l7- 0254-x), inflammatory bowel disease (doi: l0. l038/ctg.20l7.58), celiac disease

(https://doi.orgl0.H86/s40l68-0l8-04l5-6), obesity (doi: l0.3390/nu7042237), necrotizing enterocolitis (DOI 10.1186/S40168-017-0248-8)), diversity indices (e.g. richness, alpha and beta diversity), similarity to reference profiles (e.g. healthy microbiomes, microbiomes associated with particular ages, developmental stages, ethnicities, geographic locations, indoor locations [e.g. home, daycare, school, university, office, gym, nursing home, hospital, etc.], outdoor locations [e.g. park, forest, beach, ocean, river, mountain, snowy, arid, etc.), similarity to other relevant profiles (e.g. sibling, parent, classmate, colleague, celebrity, athlete, etc.), identification of populations with similar microbiome profiles, change over time, change attributable to an intervention (where intervention refers to change in diet, lifestyle, habits, medication, use of edible products, use of non-edible products, etc.), evaluation of impact of intervention, actionable biomarkers, recommendations for treatment, lifestyle recommendations, infant formula recommendations, baby food recommendations, dietary recommendations, nutritional supplement recommendations, prebiotic

recommendations, probiotic recommendations, fortified baby formula recommendations (e.g. with Human Milk Oligosaccharides, such as Similac Pro Advance, galactooligosaccharides, such as FrieslandCampina Vivinal GOS, Lactobacillus reuteri probiotic, such as Gerber Soothe, etc.), supplemented baby food (e.g. HappyBaby Organic Probiotic Baby Cereal, Nestle Cerelac, etc.), other product recommendations (e.g. diapers, bottles, fabrics, etc.), determination of product effectiveness (i.e. degree of resultant microbiome improvement or worsening), etc.

[0086] In one instance, an individual fecal sample can be analyzed on a server and an individual (e.g., end user, family of end user, physician of end user, wellness professional, coach, nutritionist, governmental body) can be virtually connected to a social network, chat group, messaging app group (e.g. Facebook Messenger, Instagram, WeChat, etc.), Facebook group etc., where the virtual group is comprised of other such individuals with similar microbiomes (in the case of the end-user) or interested in similar microbiomes (in the case of parents, health professionals, etc.), corresponding to the populations with similar

microbiomes indicated in the report.

[0087] As shown in Fig. 10, an infant can have indications of conditions (e.g., colic, eczema, or diarrhea) or discomfort. An inquiry 1 can be made to obtain annotations that can be associated with the indications. For example, annotations can include intake of microbiome disruptors (e.g., antibiotics), life events or transitions (e.g., travel, or change in day care facility). The microbiome can be characterized using the present system as shown in 2. The microbiome data can be collected and transferred to a server (local or remote) for analyzing the data. A report can be sent on a personal device (e.g., tablet) of a consumer as shown in 3. The report can include recommendations, such as intake of prebiotics. The report can include evaluation of general condition and/or health of the infant based on the microbiome. The report can direct the consumer to virtual groups, where the virtual group is comprised of other such individuals with similar microbiomes (in the case of the end-user) or interested in similar microbiomes (in the case of parents, health professionals, etc.), corresponding to the populations with similar microbiomes indicated in the report.

[0088] In some aspects, a server comprising imaging and/or microbiome data from multiple samples can be integrated within a data infrastructure (e.g.,“integrated samples”), comprising: annotations associated with user microbiome data (input by user, obtained automatically via the internet, obtained via connected devices and wearables, obtained via hospital equipment, obtained via electronic health records, etc.), a constantly updating database of published clinical trials using prebiotics, probiotics, synbiotics, human milk oligosaccharides, galactooligosaccharides, nutritional supplements, etc. indicating efficacy under certain conditions and in certain populations (e.g. DOI: 10.1177/0884533615610081, doi: 10.3945/ajcn.2008.26919, DOT 10.1097/MPG.0000000000001623, BMJ. 2001 Jun 2;322(7298): l327), a constantly updating database of products containing said substances evaluated in said clinical trials (e.g. Gerber Soothe contains Lactobacillus reuteri to combat colic, Similac Pro Advance contains Human Milk Oligosaccharides, etc.), a constantly updating database of published sequencing-based studies that determine microbiome signatures associated with particular healthy and disease states (e.g. asthma, atopy, C-section, type I diabetes, obesity, etc.), a constantly updating database of optimized array probe sequences informed or inspired by said sequencing studies database, and a constantly updating database of microbiomes and associated annotations, interventions and outcomes obtained through ongoing use of the device by multiple users and multiple samples.

[0089] In some aspects, a server comprising imaging and/or microbiome data from multiple samples can further comprise machine learning algorithms. Such machine learning algorithms can identify patterns within 2 or more temporally different samples (e.g., fecal samples) from an individual, identify patterns using a collection of samples from a plurality of individuals, identify patterns that distinguish a collection of samples representing a particular population/cohort from all other samples, and/or identify patterns that distinguish a collection of samples representing a particular population/cohort from one or more particular population(s)/cohort(s). Machine learning can also take as input annotation data regarding the individual(s) diet, drug intake, vitamin/supplement intake, physical health status, mental health status, age, ethnicity, chronic conditions, acute conditions, disease, non-medical condition (e.g. colic, diarrhea, eczema, etc.), microbiome disruptors (e.g. C-section birth, antibiotic use by end-user, antibiotic use by mother while in womb, lack of breastfeeding), life events and transitions (e.g. flying to another geographic location, starting daycare, starting solid foods, beginning to crawl, beginning to walk, moving homes, getting a dog, etc.), intervention (when applicable), time since meal, meal contents, amount of sleep, quality of sleep, time outdoors, amount of physical activity, type of physical activity, stress level, anxiety level, mood, pain level, allergy severity, physical wellbeing, mental wellbeing, consumer genomics testing results, wearables and connected device data, electronic health records, etc. and determine meaningful and/or significant associations between classes of individuals, their microbiomes and associated annotations. In an aspect, a method for analyzing microbiome-derived data of a subject can comprise using a microarray comprising probes capable of hybridizing with nucleic acid molecules to obtain the microbiome-derived data of said subject as a function of time; providing said microbiome-derived data to train a machine learning system to detect meaningful and/or significant changes in microbiomes; employing said trained machine learning system to identify changes in the microbiomes; and employing said trained machine learning system to associate said measured changes with said annotation data.

[0090] In some cases, machine learning algorithms can include Principal Component Analysis to determine the statistical variables (derived from probe signal intensities) and annotations that are collectively responsible for determining the largest differences between samples in a collection of 2 or more samples, linear regression to model the relationship between variables, logistical regression to determine the probability of a sample belonging to a certain group based on a collection of variables and annotations, decision tree (or a collection thereof called a Random Forest) to classify a collection of samples into 2 or more distinct groups based on variables and annotations, Support Vector Machine to determine the variables that best distinguish between 2 or more samples, naive Bayes classifier to determine the probability of a sample belonging to a particular classification based on associated variables and annotations, Hidden Markov Model, k-Nearest Neighbor to classify the collection of samples and associated variables and annotation into k clusters based on proximity in a multidimensional space defined by the variables and annotations, k-Means to classify the collection of samples and associated variables and annotation into k clusters based on proximity in a multidimensional space defined by the variables and annotations, hierarchical clustering to iteratively divide the collection of samples into distinct classes (top- down) or iteratively merge samples into distinct classes (bottom-up), dimensionality reducing algorithms to determine the variables and annotations most informative for classification of samples, and/or gradient boosting algorithms that combine multiple machine learning algorithms heuristically to iteratively improve the classification of samples into 2 or more classes (https://www.analyticsvidhya.com/blog/20l7/09/common-machine -learning- algorithms/).

[0091] In another aspect, machine learning algorithms can be employed to classify a sample or collection of samples obtained from a single individual and generate a report indicating the microbiome and annotation characteristics of said class, as well as provide recommendations (e.g. prebiotic, probiotic, enriched formula, supplement, diet, lifestyle, etc.) to the individual informed by interventions that proved successful for other individuals belonging to the same class.

[0092] In another aspect, machine learning algorithms can be employed to classify a sample or collection of samples obtained from a single individual and virtually connects the individual to other individuals belonging to the same class.

[0093] In another aspect, machine learning algorithm can be employed on the database of microbiomes and associated annotations to expand the database of products that have a high likelihood of being effective for a particular class of individuals.

[0094] In an aspect, machine learning algorithms can be employed on the database of microbiomes and associated annotations to expand the database of optimized array probe sequences that have a high likelihood of effectively diagnosing individuals and/or assigning them to a particular class.

[0095] An exemplary database infrastructure comprising databases and machine learning algorithms is shown in Fig. 11. A database infrastructure can include a database of optimized array probe sequences (target panel) informed by the sequencing studies database. A database can include annotations associated with user microbiome data (e.g., input by user), and/or a database of published clinical trials (e.g., using probiotics) indicating efficacy under certain conditions and in certain populations. Machine learning algorithms can be applied to the database of microbiomes and associated annotations to update the database of optimized array probe sequences that have a high likelihood of effectively diagnosing individuals and/or assigning them to a particular class. Machine learning algorithms can be employed on the database of microbiomes and associated annotations to update the database of products that have a high likelihood of being effective for a particular class of individuals.

[0096] While preferred embodiments of the present disclosure have been shown and described herein, it will be obvious to those skilled in the art that such embodiments are provided by way of example only. Numerous variations, changes, and substitutions will now occur to those skilled in the art without departing from the disclosure. It should be understood that various alternatives to the embodiments of the disclosure described herein may be employed in practicing the disclosure. It is intended that the following claims define the scope of the disclosure and that methods and structures within the scope of these claims and their equivalents be covered thereby.

[0097] Applications

[0098] The devices, systems, and methods disclosed herein can be used for a variety of applications, some of which are listed below:

[0099] At-home infant microbiome monitoring: The present system can be used for regular monitoring of the gut microbiome of an infant at home. For example, fecal samples of an infant can be monitored during different growth spurts, before and after an intervention, condition, and/or changes in environment (life events, transitions). Fecal samples can be monitored for ensuring a healthy gut microbiome of an infant. Fecal samples can be monitored for assessing effectiveness of medication or treatment.

[0100] Tracking genetically modified organisms: The present system can be used for tracking genetically modified organisms by customizing a fluidic device for processing various tissues, such as plant material and by customizing a microarray for detecting non-host sequences. For example, probes can include sequences complementary to vectors, plasmids, or other nucleic acid sequences that are commonly used in genetic engineering procedures. This can enable automatic processing of a sample followed by detection of genetically modified organisms. In some cases, processed products can be tested for genetically modified organisms. For example, soups, ketchups, chips can be tested for the presence of genetically modified corn, tomato, or other crops. In some cases, processed products can be tested for allergens from an organism. For example, processed foods can be tested for the presence of peanuts.

[0101] Metagenomics: The present system can be used to process environmental samples to relatively quantify (absolutely or relatively) presence of different microorganisms in a given ecosystem. For this application, a microarray can be customized to include probes for detecting microorganisms from a given ecosystem, such as pond, soil, etc. For example, soil microbiome can be compared from various geographic locations for identifying microbiome associated with fertile, productive soils, or soil in which crop pathogens are prevalent.

[0102] Maintenance of cell cultures: The present system can be used to verify genotype of cell cultures, by verifying mutations, transfected genes. This can be useful as a quality check during production of cell cultures.

[0103] Verification of probiotic products: The present system can be used to verify the composition of probiotics, by verifying genomic sequences (or synthetically introduced genomic barcodes) unique to each of the microbes included in the probiotic. This can be useful as a quality check during production of probiotics, to detect contaminating microbes, as well as to verify if a product includes the microbes advertised and/or at the advertised concentration.

[0104] Identification of commercial species: The present system can be used to identify different species used commercially and to detect if the product is contaminated with other species. For this application, a microarray can be customized to include probes for detection of contaminating species. For example, hamburgers can be tested using the present system for detecting low value fillers, such as soy -based fillers.

[0105] Contraband identification: The present system can be used to identify contraband in a sample. For example, a microarray can be customized to include probes for detection of contraband, such as ivory. This can be useful at immigration ports to quickly test suspected samples using the present system to determine if the suspected samples carry any contraband. This can also be useful in preserved forests where suspected samples can be quickly tested to determine if they carry any contraband.

[0106] Antibiotic resistance characterization: The present system can be used to quantify antibiotic resistance genes and their transcripts in a sample. For example, a microarray can be customized to include probes for detection of antibiotic resistance genes (using sequence databases, e.g. the McMaster University Comprehensive Antibiotic Resistance Database: https://card.mcmaster.ca/). This can be useful in NICUs, PICUs, ERs, hospitals, surgery rooms, geriatric wards, nursery homes, etc. to quickly test samples from at-risk or very ill patients using the present system to determine of their samples carry antibiotic resistance genes that could pose a danger to the patient as well as other patients in the same location. This can also be useful in the context of water supply and wastewater (in hospitals and in the community), where samples can be quickly tested to determine if they carry antibiotic resistance genes that could pose a threat to the people or animals ingesting or exposed to said water. This can also be useful in the context of foods (especially meat), where samples can be quickly tested to determine if they carry antibiotic resistance genes that could pose a threat to the people or animals ingesting said food.

[0107] Hospital/clinical diagnosis and monitoring: The present system can be used to diagnose disease, track disease progression, and evaluate therapeutic interventions. For this application, a microarray can be customized to include probes for the identification of different pathogens, antibiotic resistance genes, viruses, etc. as well as quantification of human genes and gene transcripts indicative of disease onset, progress, and remission. For example, premature babies in the NICU can be monitored for potentially life-threatening infections and antibiotic resistance, and to determine appropriate treatment to eradicate said infections and antibiotic resistance. This can also be useful in the context of surgeries and biopsies, wherein host and microbial nucleic acids derived from biopsied tissue are interrogated for copy number variation, structural variation, gene expression, relative abundance and/or other relevant biomarkers indicative of a particular cancer. This can also be useful in the context of disease diagnosis, wherein host and microbial nucleic acids derived from biopsied tissue are interrogated for copy number variation, structural variation, gene expression, relative abundance and/or other relevant biomarkers indicative of e.g. Crohn’s disease, liver cirrhosis, hepatitis C, etc.

Examples

[0108] Example 1: Representative extraction of metagenomic DNA using a microfluidic apparatus

[0109] Total bacterial genomic (metagenomic) DNA was extracted from a commercially available defined bacterial consortium using a custom, automated microfluidic chip assembly and evaluated by DNA sequencing. The ZymoBIOMICS Microbial Community Standard (ZYMO Research Corporation, Orange, CA, USA), is composed of defined amounts of three Gram+ bacteria, five Gram- bacteria, and two yeasts. (FIG. 12A).

(https://files.zymoresearch.com/protocols/_d6300_zymobiomics _microbial_community_stan dard.pdf). 120 mΐ of this cellular suspension was combined with 220ul binding buffer (Claremont Biosolutions, Claremont CA) and injected via automated multiport syringe pump into a microfluidic chip. The sample was propelled through chambers containing 1. A PCR- inhibitor removal matrix (Pre-column beads, Claremont Biosolutions, Claremont CA), 2. Garnet beads (OPS Diagnostics, Lebanon, NJ), 3. DNA binding beads (Binding Beads, Claremont Biosolutions, Claremont CA). Lysis was achieved by shaking in the presence of garnet beads for 30 minutes on a custom build shaker apparatus transmitting vibration from a 3 V micromotor (Precision Microdrives, www.precisionmicrodrives.com). Binding was achieved by incubation with binding beads, followed by washing with 500ul of stool washing buffer (Claremont Biosolutions, Claremont CA). After drying by an air push, DNA was eluted with 250ul elution buffer (Claremont Biosolutions, Claremont CA). In parallel, metagenomic DNA was extracted from the same bacterial consortium by the Qiagen

PowerSoil kit per manufacturer’s instructions Resulting DNA from both methods was sent for composition analysis by commercial 16S and ITS DNA sequencing (Microbiome

Insights, Vancouver, BC). Results (shown in FIG. 12B): DNA extracted by automated microfluidic system more closely reproduces the theoretical composition of the starting sample than the industry standard PowerSoil kit. Both easy to lyse (Gm-) and challenging to lyse (Gm+ and yeast) cell types yielded DNA at the expected concentrations, indicating an unbiased extraction methodology.

[0110] Example 2: Bacterial Community Composition Determined by Automated Microfluidic DNA Extraction and Custom Microarrav Analysis

[0111] Custom DNA microarrays were designed for analysis of the defined bacterial consortium described above. Genome sequences were obtained for the 10 organisms present, as well as 5 organisms that are not present in the sample (Bifidobacterium infantis,

Bifidobacterium longum, Bifidobacterium breve, Klebsiella pneumonia and Clostridium botulinum). Candidate 40mer oligonucleotide sequences were generated for each genome, filtered for thermodynamic criteria including %GC and Tm, and evaluated for cross- hybridization to a background set of 600 common human commensal organisms. Probes which passed threshold criteria were downselected for optimal Tm=80°C. Approximately 100 sequences per genome were selected as the test probe set. Each probe was printed in replicated (n=5) on a 7500-feature custom 6-pack microarray (Arbor Biosciences, Ann Arbor, MI), along with conventional negative, positive, spike in and stringency control features. DNA from the defined bacterial consortium was extracted using the automated microfluidic system described in example 1. Additionally, commercially extracted DNA was obtained from the manufacturer (ZymoBIOMICS Microbial Community DNA Standard, Zymo Research) representing the same theoretical community composition. For each, 2ug of DNA was labeled with Cy3 using primer extension (CyTag CGH labeling kit, Enzo Life Sciences) according to manufacturer’s instructions. Labeled DNA was cleaned up using Agilent SureTag DNA Labeling Kit Purification Columns according to manufacturer’s protocol lug of labeled DNA was combined with 2ul salmon sperm DNA (Invitrogen, lOug/ul), lul control spike-in targets (Arbor BioSciences), and brought to 27ul with water, then combined with 27ul 2X Hi-RPM Hybridization Buffer (Agilent Technologies) and applied to the microarray. Hybridization was carried out in a rotating oven for 24hrs at 60°C, 20rpm. Arrays were washed in Agilent CGH Wash buffers 1 (5 min RT) and 2 (3 min 50°C) and scanned on an Axon GenePix 4000B microarray scanner (Molecular Devices). Data was extracted using GenePix 6.0 software. The median Cy3 foreground-background signal was calculated per feature. Feature replicate outliers were removed, and remaining replicates features were averaged. The median signal for all probes representing a given organism were calculated. Results: Average signal levels for probes that correspond to organisms not present in the control sample were lower than for those organisms present in the sample, as shown in FIG.

13 A. Average signal intensity correlated to expected genome abundance within an order of magnitude. Commercially extracted DNA resulted in higher average signal levels for Gram+ organisms than for Gram- organisms, as shown in FIG. 13B, whereas less bias was observed for DNA extracted by automated microfluidic system (FIG. 13 A).

[0112] Example 3: Configuration of a fluidic cartridge

[0113] FIG. 14 shows a configuration of a fluidic cartridge 1400. As shown in the figure, the fluidic cartridge 1400 comprises fluidic processing elements and a microarray. The fluidic cartridge comprises a plurality of reaction chambers in communication with reagents for extracting and labeling nucleic acid molecules for further nucleic acid analysis using a microarray. In this configuration, Dissolution Reagent in 1 is in communication with Input Reaction Chamber 6 via a microchannel. Once the cartridge is actuated by addition of a sample, Dissolution Reagent flows into Input Reaction Chamber 6 to form a slurry. Input Reaction Chamber 6 is also in communication with Waste Chamber 9 for releasing undissolved sample into Waste chamber 9. The slurry is transferred to NA Extraction

Reaction Chamber 7 via a microchannel for extracting nucleic acid molecules. Lysis reagent in 2 is released into NA Extraction Reaction Chamber 7 for lysing cells to extract nucleic acid molecules. Wash reagent 3 is released into NA Extraction Reaction Chamber 7 to wash the extracted nucleic acid molecules. Reagents from NA Extraction Reaction Chamber 7 flow to Waste Chamber 9 for storage. The extracted nucleic acid molecules are transferred to NA Labeling Reaction Chamber 8 via a microchannel. NA Labeling Reaction Chamber 8 is provided with Elution Reagent 4 to elute bound nucleic acid molecules and to label the eluted nucleic acid molecules with Labeling Reagent 5 (optionally including spike-in control oligonucleotides that would be labeled simultaneously). Reagents from NA Labeling

Reaction Chamber 8 are flown to Waste Chamber 9. The labelled nucleic acids from NA Labeling Reaction Chamber 8 are transferred to Microarray 10 for hybridization with probes. The fluidic cartridge 1400 is also equipped with Identification tag 11.