METHODS AND COMPOSITIONS FOR IMPROVING COFFEE

FLAVOR

CROSS REFERENCE TO RELATED APPLICATIONS

[0001] The present application claims priority to US Provisional Application No.

61/887,890, filed October 7, 2013, and US Provisional Application No. 61/949,872, filed March 7, 2014, the disclosures of which are incorporated in their entireties herein.

BACKGROUND OF THE INVENTION

[0002] The two main varieties of coffee grown in the world are Arabica (Coffea arabica) and Robusta (Coffea canepohora). These plants are members of the fourth largest family of flowering plants, the Rubiaceae, which contains 611 genera and 13,900. Rubiaceae are widespread, but somewhat more abundant in warmer and tropical climates around the world. Robusta coffee is relatively free from pests and while Arabica coffee is more valuable commercially, it has a greater variety of pests.

[0003] Nearly all coffee in Rwanda, Burundi and Uganda is grown on small holdings. Antestia bugs are the most serious pests of Arabica coffee in the Great Lakes region of East Africa, but are not present elsewhere. Antesia bugs (genus Antestiopsis) damage growing points of coffee plants, and are associated with potato taste defect (PTD). PTD is usually detected after the coffee beans are roasted and brewed, but it can also be detected earlier by smell in severe cases. Once a crop is found to have PTD, it becomes worthless on the market. While PTD represents a specific example of a flavor defect, other coffee-growing regions have flavor defects associated with their specific conditions, flora, and fauna.

[0004] PTD in processed coffee beans is strongly associated with antestia, which physically transmit microbial pathogens between crops when feeding. Since roasting of coffee kills all microbes present, the PTD chemical(s) survive or are chemically altered during roasting to contaminate the final ground coffee. [0005] The coffee berry borer (CBB), Hypothenemus hampei, which originates in Central Africa, is another serious pest. CBB are often affected by Wolbachia bacteria. Climate change has already resulted in increased damage from coffee berry borer and has created a wider distribution of the pest. Projections point to an increased loss in coffee production in Rwanda and the other countries of the East African community especially for Arabica coffee production areas. While CBB is responsible for significant loss of coffee cherries, the damaged beans are easily eliminated in processing, and CBB is not associated with taste defects like those attributed to antestia complex and its associated microbes.

[0006] Losses due to PTD are estimated to amount to 30- 40% of the crop in Rwanda and Barundi. While fenitrothion is suitable for control of insect pests, farmers in emerging economies cannot afford expensive sprays. Insecticide application can create new insect problems, or select for resistance to insecticides, and is increasingly disfavored on the market. Recent trends in the Rwanda and Barundi coffee industry are towards more organic methods. However, a biological control infrastructure is missing in much of Africa, and starting one would require commitment of personnel plus collaboration or outside training.

BRIEF SUMMARY OF THE INVENTION

[0007] Provided herein are methods and compositions to eliminate flavor defects, and improve flavor profile, of coffee prior to roasting. These methods can be carried out at the point of processing, e.g. , after export, and thus do not require infrastructure in coffee-growing regions. The presently described methods address the waste and financial loss associated with coffee flavor defects, and can be applied to non-defective coffee beans to improve flavor.

[0008] Provided herein are methods of killing microbes on raw coffee beans comprising exposing the beans to sterilization, thereby killing the microbes. In some embodiments, the sterilization is via ultraviolet (UV) irradiation. In some embodiments, the method further comprises washing the beans to remove microbial residue and optionally drying the beans. In some embodiments, the method further comprises preserving liquid used to wash the beans to confirm efficacy of microbial killing, e.g., by culturing the liquid, centrifuging the liquid, viewing the liquid microscopically, etc. In some embodiments, the raw coffee beans are known to have a flavor defect (undesirable flavor), e.g., potato taste defect (PTD). In some embodiments, the raw coffee beans carry Staphylococcus -related sp., Leuconostocacea, and/ or Aspergillus prior to sterilization/ irradiation.

[0009] In some embodiments, the UV irradiation is from a UV source/ device having at least 50 watts, e.g., 200, 250, 300, 500, or higher power UV source. In some embodiments, the exposing is for at least 1 minute, e.g., 2, 5, 10, 20, 30, 40, 50, 60, 2-10, 5-20, 20-60, 60- 90, 20-120, or more minutes. In some embodiments, the exposing is in a basket chamber or surface capable of agitating or rotating the raw coffee beans to ensure complete exposure of the beans to UV light. In some embodiments, the exposing includes UV irradiation from a reflective surface, e.g., metal, mirror, foil, etc. For example, the basket chamber or surface is exposed to the reflective surface such that the UV irradiation comes from the source, as well as the refective surface. In some embodiments, the UV source emits UV-C.

[0010] In some embodiments, the method further comprises adding favorable microbes to the beans. In some embodiments, the favorable microbes are from beans with desirable flavor. In some embodiments, the favorable microbes out-compete non- favorable microbes, e.g., those associated with undesirable flavor. In some embodiments, the favorable microbes are applied in a liquid composition.

[0011] In some embodiments, the exposing results in an increased Sensorial Analysis score of the beans compared to the Sensorial Analysis score of beans from the same sample or region that are not exposed to irradiation. For example, the exposing can result in a Sensorial Analysis score of at least 78, 79, 80, 81, 82, 83, 84, 85, 80-90, 80-95 or higher. In some embodiments, the Sensorial Analysis is performed after the beans are roasted and brewed. In some embodiments, the Sensorial Analysis is performed without addition of favorable microbes. In some embodiments, the Sensorial Analysis is performed after addition of favorable microbes.

[0012] Provided herein are methods of obtaining favorable microbes for modifying {e.g., improving or standardizing) flavor of coffee obtained from a first sample of raw coffee beans comprising: (i) obtaining a second sample of raw coffee beans, wherein the second sample of raw coffee beans carries favorable microbes; (ii) washing the second sample of raw coffee beans in liquid; and (iii) separating the second sample of raw coffee beans from the liquid. In some embodiments, the method further comprises incubating the liquid separated in step (iii). In some embodiments, the incubating is for at least 1 day, 3 days, 1-7 days, 1 week, 7-30 days, 1 month, or more. In some embodiments, the washing comprises soaking the second sample of raw coffee beans for 1-30 minutes, e.g., 2, 5, 10, 20, 30, 40, 50, 60, 2-10, 5-20, 20- 60, or more minutes. In some embodiments, the liquid is concentrated and/or frozen

(cryopreserved) after incubating (culturing). [0013] Further provided are compositions for modifying the flavor of coffee obtained from a first sample of raw coffee beans, wherein the composition comprises liquid used to wash a second sample of raw coffee beans carrying favorable microbes. Also provided are compositions for increasing the amount of favorable microbes on raw coffee beans comprising liquid used to soak or wash raw coffee beans with a desirable flavor. In some embodiments, the liquid is formulated into a composition for application to raw coffee beans. In some embodiments, the liquid is cultured (incubated) after removal of the raw coffee beans with a desirable flavor (e.g., the second sample of raw coffee beans) from the liquid, e.g., for 1 day, 3 days, 1-7 days, 1 week, 7-30 days, 1 month, or more. In some embodiments, the composition is concentrated and/or frozen (cryopreserved) after removal of the raw coffee beans with a desirable flavor from the liquid. In some embodiments, the composition is concentrated and/or frozen (cryopreserved) after further culturing. In some embodiments, the raw coffee beans with a desirable flavor (e.g., second sample of raw coffee beans) are washed for at least one minute, e.g., 2, 5, 10, 20, 30, 40, 50, 60, 2-10, 5-20, 20-60, or more minutes. [0014] Further provided are methods for modifying the flavor of coffee obtained from a first sample of raw coffee beans comprising: adding to the first sample of raw coffee beans a composition comprising liquid used to wash a second sample of raw coffee beans carrying favorable microbes (e.g., coffee beans known to produce coffee with a desirable flavor), thereby modifying the flavor of coffee obtained from the first sample of raw coffee beans. In some embodiments, the first sample of raw coffee beans are sterilized, e.g. , exposed to UV irradiation, prior to said adding. In some embodiments, the first sample of raw coffee beans are known to have a flavor defect (undesirable flavor), e.g., PTD. In some embodiments, the first sample of raw coffee beans carry Staphylococcus -related sp., Leuconostocacea, and/ or Aspergillus prior to sterilization/ irradiation. In some embodiments, the first sample of raw coffee beans are not associated with a flavor defect. In some embodiments, the method further comprises roasting the first sample of raw coffee beans after said adding.

[0015] In some embodiments, the composition is formulated for application to raw coffee beans, e.g., in a liquid for spray, dusting, wash, or soak. In some embodiments, the liquid is cultured (incubated) after removal of the second sample of raw coffee beans from the liquid, e.g., for 1 day, 3 days, 1-7 days, 1 week, 7-30 days, 1 month, or more. In some embodiments, the composition is concentrated and/or frozen after removal of the second sample of raw coffee beans from the liquid. In some embodiments, the composition is concentrated and/or frozen after further culturing. In some embodiments, the second sample of raw coffee beans are washed for at least one minute, e.g., 2, 5, 10, 20, 30, 40, 50, 60, 2-10, 5-20, 20-60, or more minutes.

[0016] In some embodiments, the addition of favorable microbes result in an increased Sensorial Analysis score compared to the Sensorial Analysis score of beans from the same sample or region without added favorable microbes. For example, the addition can result in a Sensorial Analysis score of at least 78, 79, 80, 81, 82, 83, 84, 85, 80-90, 80-95 or higher.

[0017] Also provided are apparatuses for killing microbes on raw coffee beans, wherein an apparatus comprises: a basket chamber or surface capable of agitating or rotating the raw coffee beans; and a sterilization source positioned to expose the raw coffee beans to sterilization. In some embodiments, the sterilization source is a UV source. In some embodiments, the UV source has at least 50 watts, e.g., 100-500, 200, 250, 300, 500, 1000, or higher power. In some embodiments, the UV source emits UV-C. In some embodiments, the apparatus comprises a rotating basket chamber. In some embodiments, the apparatus comprises an agitating surface. In some embodiments, the apparatus has a capacity for at least 0.5 pounds of raw coffee beans, e.g., 1, 2, 5, 50-20, 20-100, 100-250, 10, 20, 50, 100, 250, or more pounds of raw coffee beans. In some embodiments, the apparatus further comprises at least one reflective surface, i.e., positioned to reflect and amplify the UV light on the beans. In some embodiments, the reflective surface surrounds the chamber or surface. In some embodiments, the reflective surface is positioned opposing the UV source in the apparatus. In some embodiments, the reflective surface is selected from the group consisting of foil, metal, and mirror.

BRIEF DESCRIPTION OF THE DRAWINGS

[0018] Figure 1. Three different fungal subcultures taken from the potato taste defect (PTD) SG-1416 plate (top center). Fungal isolates (left, right and bottom) were allowed to grow for a week. Circles and labeling on the original plate (top center) indicate where each sample was taken.

[0019] Figure 2. Microbial colonies taken from a control bean (SG-1423, left) and a PTD bean (SG-1419, right) grown on potato dextrose agar (PDA) medium. The dashed line separates the area where the rear (R) or front (F) of the bean was swiped.

[0020] Figure 3. Image of PDA (potato dextrose agar) culture plate of a PTD bean (SG- 3238) after six days of incubation at 21C. The arrows indicate the slate blue fungi cultures overtaking bacteria samples that have grown previously. Both sides of the bean contain the same species of bacteria.

[0021] Figure 4. Flasks of water used to soak UV-irradiated coffee beans with PTD (SG- 3160) after one week of culture. The beans were UV-C irradiated for the indicated time (5, 10, 15, 20, 25, 30, 40 or 50 minutes) and added to sterile water for 20 minutes. The water and beans were then separated, and microbe growth observed in the water at 21C. For the 50 minute UV sample, the soak water remained microbe-free at one week and remained so for at least two months.

[0022] Figure 5. Growth of bacterial and fungal colonies after 3 days of culture at 21C. Samples were taken from the liquids shown in Figure 4 for the 5-30 minute samples and plated PDA.

[0023] Figure 6. Growth of bacterial and fungal colonies after one month of culture for the 5-20 minute samples.

[0024] Figure 7. An embodiment of the rotating drum for UV irradiation is shown with a reflective foil lining.

[0025] Figure 8. Figure 8 A shows a fan for cooling the UV lamps associated with the irradiating chamber. Figure 8B shows an embodiment of the irradiating machine, with a hose connecting the fan to the chamber, and a barrier erected in front of the chamber to reduce potential contamination. [0026] Figure 9. Taste profile and scores of individual tasters for Burundi #5 coffee beans with no treatment.

[0027] Figure 10. Taste profile and scores of individual tasters for 1 pound Burundi #5 coffee beans after 6 hours UV treatment and fan.

[0028] Figure 11. Taste profile and scores of individual tasters for 1 pound Burundi #5 coffee beans after 3 hours UV treatment, fan, and aluminum reflective covering.

[0029] Figure 12. Taste profile and scores of individual tasters for 1 pound Burundi #5 coffee beans after 6 hours UV treatment, fan, and aluminum reflective covering.

[0030] Figure 13. Taste profile and scores of individual tasters for 0.5 pound Burundi #5 coffee beans after 6 hours UV treatment, fan, and aluminum reflective covering. DETAILED DESCRIPTION OF THE INVENTION

Introduction

[0031] Provided herein are methods and compositions to remove undesirable flavor associated with microbes from coffee beans prior to roasting. Additionally provided are methods and compositions to add desirable flavor associated with microbes to coffee beans prior to roasting. Neither UV irradiation of non-roasted beans, or a "probiotic" approach to improving flavor have been tried before. UV irradiating beans prior to roasting kills microbes associated with the beans before residue from the microbes can be "baked into" the beans during roasting. This approach is useful for the potato taste defect (PTD) found in

Africa, but can be applied to coffee from any region where region-specific taste defects arise. Similarly, adding microbes from beans with a known, desirable flavor profile to other beans, whether irradiated or not, results in transfer of the good flavors associated with microbial residue to the other beans. The techniques can be used in tandem where beans are exposed to UV light, optionally checked to verify microbial killing, and then treated with favorable microbes prior to roasting.

Definitions

[0032] Unless defined otherwise, technical and scientific terms used herein have the same meaning as commonly understood by a person of ordinary skill in the art. See, e.g., Lackie, DICTIONARY OF CELL AND MOLECULAR BIOLOGY, Elsevier (4 ^TH ed. 2007); Sambrook et al , MOLECULAR CLONING, A LABORATORY MANUAL, Cold Springs Harbor Press (Cold Springs Harbor, NY 1989). Any methods, devices and materials similar or equivalent to those described herein can be used in the practice of this invention. The following definitions are provided to facilitate understanding of certain terms used frequently herein and are not meant to limit the scope of the present disclosure.

[0033] The term "microbe" refers to single-cell organisms such as bacteria, archaea, protists, and fungi. The term "favorable microbe" in the context of the present disclosure refers to microbes that: impart a desirable flavor when roasted on coffee beans, outcompete or otherwise reduce numbers of non-favorable microbes. A "non-favorable microbe" is one that, when present on a coffee bean during roasting, causes an undesirable flavor. For potato taste defect (PTD), non-favorable microbes include bacterial Staphylococcus -related sp. and Leuconostocacea sp., fungal Aspergillus sp., gammaproteobacterium, Spiroplasma, Sodalis, and Rickettsia.

[0034] The terms "raw," "uncooked," "green," refer to coffee beans prior to roasting.

Methods of microbial detection

[0035] Microbes can be detected using methods known in the art. Such methods can be used, e.g. , to confirm complete microbial killing after sterilization, e.g. , UV irradiation, where complete killing is confirmed by lack of microbial growth. Such methods can also be used to detect the identity of microbes associated with flavor defects or desirable flavors. [0036] For example, a sample can be taken, e.g. from a coffee bean or liquid used to wipe or soak the bean. The sample can then be subjected to standard culturing. Alternatively, the presence and/or identity of the microbe can be determined by genetic or protein profiling of the sample. For example, a particular microbe can be detected by detecting DNA or R A {e.g. 16S rRNA) sequence or protein known to be associated with the microbe. NCBI hosts a database with genomic information for thousands of microbial species {see, e.g., the website at ncbi.nlm.nih.gov/genome/browse). A number of kits for microbial detection are commercially available, e.g., from Life Technologies®, Waterworks®, Cole-Parmer®, etc.

[0037] The following discussion of the invention is for the purposes of illustration and description, and is not intended to limit the invention to the form or forms disclosed herein. Although the description of the invention has included description of one or more embodiments and certain variations and modifications, other variations and modifications are within the scope of the invention, e.g., as may be within the skill and knowledge of those in the art, after understanding the present disclosure. All publications, patents, patent applications, Genbank numbers, and websites cited herein are hereby incorporated by reference in their entireties for all purposes.

Examples

[0038] We sought to discover what causes the potato taste defect (PTD) in coffee and determine how to eliminate the causative agent from raw coffee beans. Ideally, this could be done before or after export, at low cost, and not sacrifice desired flavors in the coffee. Example 1: Identification of microbes associated with PTD

[0039] PTD is strongly associated with feeding damage by shield bugs of the genus Antestiopsis, called collectively antestia. Antestia are known to carry microbial pathogens. The roasting process kills all microbes present, so the PTD chemical(s) from microbes survive or are chemically altered during roasting to contaminate the final ground coffee.

[0040] Offending PTD microbes can be shipped with raw green coffee beans to roasters, a trip than can take 2-4 months. When raw PTD coffee beans are wiped on potato dextrose agar (PDA) culture plates, a mixture of bacterial and fungal colonies grow, with bacteria typically growing faster than fungi. The plated colonies have distinctive odors. [0041] To determine which microbes are associated with PTD, we wiped single beans from selected control (non-PTD) and PTD samples on PDA culture plates. The plates were incubated at 21° C. Colonies from control (non-PTD) beans were markedly different from colonies from PTD beans when observed for the week following plating.

[0042] Fungal colonies from control beans initially were slow to develop, but over the week, the fungal colony growth well outpaced the bacterial colonies and eventually over grew them. In the colonies from PTD beans, bacterial growth was much more prevalent. The result indicates that bacterial growth is associated with PTD, and that a dynamic interaction between bacterial and fungal colonies exists.

[0043] An olfactory test was also conducted after the plates were allowed to mature for one week. This test was a general test that sensed which microbes were present on an individual bean. A combination of the different fungal and bacterial samples contributed to an individual scent that was recorded. After smelling and recording the results of each plate, olfactory senses were cleansed by smelling ethanol and resuming the test again after one minute to guarantee that the previous scent would not affect the discernment of the next. Professional coffee tasters verified that the samples designated "control" were not affected by the potato taste defect while those with the label of "experiment" were afflicted with the defect. Fungal scents such as potato and earth notes were prevalent among the control beans while those indicating urine or mildew was indicative of excessive bacterial growth.

[0044] Each of the microbes have a distinctive smell. Isolates of each of the different bacterial and fungal specimens were taken in order to verify this. Figure 1 illustrates the diverse fungal colonies present on just one PTD coffee bean. [0045] A fungus of note is shown on the far right of Figure 1. It is slate blue in color, and was prevalent on all the control plates, and to a lesser extent on the PTD plates. It appears to displace bacterial samples and outgrow other fungal species when present in the same culture dish. During a streak replating of the culture, mycelia emerge from the streaks where the spores were placed on the first day after plating, and spores start to emerge on the second day. During the scent test, its scent was also shown to increase proportionally to the number of spores present. This is in contrast to the black fungus shown in the bottom center of Figure 1, in which scent did not correlate with the number of spores.

[0046] The dramatic difference in appearance between microbial colonies from control beans compared to PTD beans is indicated in Figure 2. The evidence indicates that one or more of the fungi are producing antibiotics to degrade or otherwise overpower the bacterial colonies.

[0047] Roasting and plating of the resulting beans was also conducted. Beans with the potato taste defect were roasted in a Behmor 1600 roasting oven for seven minutes and forty- five seconds, with the first crack occurring at seven minutes and thirty seconds. The beans were roasted at the setting meant for a quarter pound of coffee within the cycling drum. The beans were then allowed to cool for ten minutes after the roasting. The total cook and cool time for the beans was an average of 20 minutes and nineteen seconds per batch of beans. Approximately 156 grams of the beans were roasted at one time from one batch. After roasting, the beans ranged in color from city roast to full city roast. They weighed approximately 101 grams after being allowed to cool, indicating desiccation during the roasting process. The subsequent roasted beans were then soaked in autoclaved distilled water and plated just like their raw counterparts. When examined daily, the samples showed no bacterial or fungal presence after two weeks. However, on tasting the roasted coffee, there was still evident PTD. This is due to residual chemicals biosynthesized by the microbes before roasting, though roasting may have modified somewhat.

[0048] Samples of microbes from the beans revealed the presence of Staphylococcus - related and Leuconostocacea bacteria, and Aspergillus fungi. Because antestia insects found on the beans are known to carry microbes, we also sequenced 16S rRNA for microbes found on insects in the area. This revealed the presence of gammaproteobacterium, Spiroplasma, Sodalis, and Rickettsia symbionts. Example 2

[0049] We set out to eliminate microbes on affected coffee beans using UV light. The PTD (SG- 3160) coffee bean batch was used for this test. The particular light used was a 44 Watt, 120 volt UV-C bulb, but for commercial scale and time pressures, a more powerful bulb will be used (e.g., 200, 250, 300, 350, or 500W bulb). We used a traditional roaster (Behmore 1600 Gourmet Coffee Roaster) with a basket cylinder made of stainless steel mesh of square 4 mm openings on a reinforced frame. There are small vanes attached inside designed to constantly turn the bean over and the cylinder is rotated during roasting, or in this case, irradiating. The cylinder has latches and hinges on the end to open and insert green coffee beans. The cylinder is 26 cm long and 12.5 cm in diameter, and can accommodate a pound of coffee beans according to the Behmor instruction manual. The monitoring lamp was removed and replaced with the UV bulb. Again, for commercial scale, larger cylinders, or a large flat mesh can be used to accommodate larger volumes.

[0050] Testing involving UV-C light showed that sterile water used to soak the PTD beans for 20 minutes included fungi or bacteria (shown by colored spots on surface for fungi or murkiness for bacteria). The times indicated in Figure 4 are for minutes the coffee beans were exposed to UV light. Again, with a more powerful UV bulb, the time required to kill microbes will be reduced. The flasks were cultured for one week at 21C after the beans were removed. Bacteria generally emerged at approximately two days while fungi took around three. With 50 minutes of irradiation, the soak water remained entirely clear after one week, and for several months. The flecks visible in the 50 minute sample are small amounts of the silverskin of the coffee beans.

[0051] Figure 5 shows PDA plates cultured at 21C for 3 days after plating with the liquid samples (from Figure 4) up to the 30 minute mark. Fungal and bacterial colonies are evident. Figure 6 shows the cultures up to the 20 minute mark grown for one month. The fungi and bacteria present in the plates have had time to interact. The presence of fungi is increasingly unlikely as the period of time for UV exposure increases, as they are easier to destroy than bacteria.

Example 3 [0052] Coffee bean taste can be scored on a 100 point Sensorial Analysis scale, with points given for different characteristics and flavors as follows: fragrance (Fr); taste (Ta); aftertaste (Af); acidity (Ac); body (Bo); uniformity (Un); clean cup (CI); balance (Ba); sweetness (Sw); and overall (Ov). Typically, a score of around 80 or more is required for suitability. If PTD is detected, a score of 73 is given without further characterization.

[0053] Coffee beans with PTD were received from Burundi (Burundi #5) and given a score of 73. The beans were treated UV light as described herein so they could be characterized and compared to samples without sterilization.

[0054] The device used to sterilize the beans is shown in Figure 7. Foil or some other reflective material can be used to reflect UV light. A rotisserie oven (13in x 13in x 13in) with a motor is used to rotate the metal drum holding the beans, and a blower unit is used to cool the UV quartz lights, and a box (e.g., 23in x 23in x 20in) can be used to keep out potential contaminants (see Figure 8 A and 8B). There are two UV-C wavelength lights in the oven, one fixed on the top and another on the bottom. In the present example, the top lamp is 120 volts with a 0.37 amp current while the bottom lamp is 120 volts with a 1.15 amp current.

[0055] The sterilization conditions varied as follows:

UV-C for 6 hours with a fan;

UV-C for 3 hours with reflective foil and a fan; and

UV-C for 6 hours with reflective foil and a fan.

[0056] A sample was taken from the beans from each condition, roasted, tasted, and scored (where appropriate) by multiple tasters. The results are shown below, and indicate that a reflective covering for the rotating drum increases sterilization efficiency.

[0057] Treatment with UV light in a reflective chamber resulted in no microbial growth after 1 week, even with 3 hours treatment. Additional ways to increase sterilization include increasing the strength of UV light, or the reducing the amount of coffee beans in the chamber, so that each bean is more exposed.

Example 4

[0058] Probiotic microbes have been isolated from Rwandan beans with positive flavors. The isolated microbes include blue, tan, black, brown, and green fungi. The fungi are suspended as a liquid culture in Potato Dextrose Broth to permit easier handling and to reduce the risk of contamination. Parts of the parent sample are grown in different vials to prepare for inoculation for probiotic treatment. This involves purification of the probiotic culture, sterilizing raw coffee beans; a flavoring the sterilized coffee beans with the probiotic culture, and roasting the beans to induce release of positive-flavored chemicals on the beans.