This document presents the results of statistical analysis on symptoms from viable, self-annotated Biomesight microbiome samples. The methodology for data acquisition is outlined in New Standards for Microbiome Analysis?.​

Tables have been refined to display only genus- and species-level taxa, the 20 most prominent entries per group, and associations achieving statistical significance (P < 0.01).​

The following sections provide the processed data, accompanied by guidance on interpretation and application. Counts of significant bacterial taxa are included, reflecting the application of non-standard but rigorously validated statistical approaches to extensive sample and reference populations, where statistical power derives from dataset scale.

Significance	Genus
p < 0.01	177
p < 0.001	160
p < 0.0001	138
p < 0.00001	124

Averages and Medians

I prefer medians over averages. Medians are the values where half of the people have less and half has more. If the data was a bell-curve, then the values will almost be the same… with bacteria that happens rarely. Look at Faecalibacterium prausnitzii is below, we see that the average is above and the median below.

If symptom median is higher than reference median, it means there is more of this bacteria. If lower, then less. This ignores how often the bacteria is seen (we average only over reports).

tax_name	Rank	Symptom Avarage	Reference Average	Symptom Median	Reference Median
Faecalibacterium prausnitzii	species	10.366	12.277	11.415	9.08
Faecalibacterium	genus	10.881	12.843	12.131	9.826
Lachnospira	genus	2.401	2.738	1.9	1.418
Coprococcus	genus	1.071	1.443	0.737	0.428
Phocaeicola dorei	species	2.699	2.916	0.418	0.128
Parabacteroides	genus	2.385	2.634	1.724	1.989
Clostridium	genus	2.005	1.854	1.359	1.6
Roseburia faecis	species	0.951	1.215	0.576	0.457
Bacteroides caccae	species	1.59	0.852	0.286	0.402
Mediterraneibacter	genus	0.805	0.713	0.277	0.381
Bacteroides thetaiotaomicron	species	1.104	1.071	0.463	0.561
Lachnospira pectinoschiza	species	0.547	0.667	0.336	0.249
Bifidobacterium	genus	0.761	0.94	0.127	0.042
Bacteroides cellulosilyticus	species	1.396	0.836	0.076	0.151
Blautia wexlerae	species	0.869	0.569	0.314	0.386
Bilophila	genus	0.343	0.35	0.21	0.278
Anaerotruncus	genus	0.284	0.184	0.136	0.203
Akkermansia muciniphila	species	2.398	1.325	0.05	0.117
Akkermansia	genus	2.398	1.325	0.051	0.117
Anaerotruncus colihominis	species	0.259	0.173	0.133	0.198

Bacteria Incidence – How often is it reported

The common sense belief is that if a bacteria is reported more often, then the amount should be higher. This is often not true. The microbiome is a complex thing. Look at Bacteroides uniformis below, we see that the average is above and the median below

tax_name	Rank	Incidence Odds Ratio	Chi2	Symptoms %	Reference %
Actinobacillus porcinus	species	0.61	6.9	24.5	40.2
Slackia faecicanis	species	1.53	7.8	44.8	29.2
Mogibacterium vescum	species	1.79	11.5	32.2	18

More or Less often based on Symptom Median All Incidence

This is a little more complex to understand. If we compute the mid point for people with the symptom, then if the bacteria was not involved then half of the reference should be above this value and half below this value. If not, it means that the symptom tends to over or under growth.

Psychrobacter glacialis	species	0.002	0.37	30.5	664	247
Niabella	genus	0.002	0.39	27.4	583	226
Thermoanaerobacterium	genus	0.002	0.4	24.4	485	195
Chromatium	genus	0.002	0.41	24.2	508	206
Chromatium weissei	species	0.002	0.41	24.1	507	206
Thermoanaerobacterium islandicum	species	0.002	0.41	23.6	478	195
Syntrophomonas sapovorans	species	0.002	0.42	22.5	536	226
Sporosarcina pasteurii	species	0.002	0.42	21.9	440	184
Thermodesulfovibrio thiophilus	species	0.002	0.43	21	543	236
Sporosarcina	genus	0.002	0.43	20.6	444	191
Oenococcus	genus	0.002	0.45	20.1	609	272
Thermodesulfovibrio	genus	0.002	0.45	19.5	629	285
Helicobacter suncus	species	0.002	0.47	18.3	768	361
Desulfofundulus	genus	0.002	0.46	18.2	496	227
Herbaspirillum magnetovibrio	species	0.002	0.51	13.6	447	226
Streptococcus infantis	species	0.003	0.54	12	804	437
Sphingomonas	genus	0.002	0.53	11.9	457	242
Desulfotomaculum defluvii	species	0.003	0.56	11.3	1022	570
Alkalibacterium	genus	0.003	0.57	10.5	906	514
Hydrogenophilus	genus	0.003	0.58	10.2	1149	662

More or Less often based on Reference Median All Incidence

This is like the above, but with a different line in the sand. Instead of the median of those with the condition, we use the median of the reference set.

tax_name	Rank	Reference Median	Odds Ratio	Chi2	Below	Above
Methylonatrum	genus	0.005	0.35	374.2	1861	655
Methylonatrum kenyense	species	0.005	0.35	374.2	1861	655
Anaerotruncus colihominis	species	0.198	0.41	365.1	2712	1113
Anaerotruncus	genus	0.203	0.42	340.3	2688	1141
Odoribacter denticanis	species	0.006	0.41	291	1856	760
Luteolibacter	genus	0.017	0.38	245.6	1225	468
Luteolibacter algae	species	0.017	0.39	240.4	1214	468
Finegoldia	genus	0.0115	0.41	212.1	1210	501
Anaerococcus	genus	0.012	0.4	206.3	1099	444
Eggerthella sinensis	species	0.006	0.44	197.2	1289	568
Finegoldia magna	species	0.008	0.4	195.4	1014	408
Coprococcus	genus	0.4285	1.87	191.4	1379	2577
Desulfovibrio fairfieldensis	species	0.0395	0.4	175.1	868	347
Mogibacterium	genus	0.023	0.54	170.4	2154	1159
Bifidobacterium	genus	0.04245	1.8	169.2	1390	2505
Rubritalea	genus	0.004	0.43	168.7	969	415
Bifidobacterium longum	species	0.016	1.9	167.6	986	1876
Lysobacter	genus	0.004	0.36	164.9	657	236
Porphyromonas	genus	0.013	0.54	164.7	2156	1174
Psychrobacter glacialis	species	0.002	0.37	158	664	247

More or Less often based on Symptom Median High Incidence

Above we see that many of the top bacteria identified are sparse, that is not reported often. We then restrict them to those that occur above 50% or the time.

tax_name	Rank	Symptom Median Freq	Odds Ratio	Chi2	Below	Above
Clostridium taeniosporum	species	0.003	0.6	8.6	1359	821

More or Less often based on Reference Median High Incidence

Above we see that many of the top bacteria identified are sparse, that is not reported often. We then restrict them to those that occur above 50% or the time.

tax_name	Rank	Reference Median Freq	Odds Ratio	Chi2	Below	Above
Methylonatrum	genus	0.005	0.35	374.2	1861	655
Methylonatrum kenyense	species	0.005	0.35	374.2	1861	655
Anaerotruncus colihominis	species	0.198	0.41	365.1	2712	1113
Anaerotruncus	genus	0.203	0.42	340.3	2688	1141
Odoribacter denticanis	species	0.006	0.41	291	1856	760
Luteolibacter	genus	0.017	0.38	245.6	1225	468
Luteolibacter algae	species	0.017	0.39	240.4	1214	468
Finegoldia	genus	0.0115	0.41	212.1	1210	501
Anaerococcus	genus	0.012	0.4	206.3	1099	444
Eggerthella sinensis	species	0.006	0.44	197.2	1289	568
Finegoldia magna	species	0.008	0.4	195.4	1014	408
Coprococcus	genus	0.4285	1.87	191.4	1379	2577
Desulfovibrio fairfieldensis	species	0.0395	0.4	175.1	868	347
Mogibacterium	genus	0.023	0.54	170.4	2154	1159
Bifidobacterium	genus	0.04245	1.8	169.2	1390	2505
Rubritalea	genus	0.004	0.43	168.7	969	415
Bifidobacterium longum	species	0.016	1.9	167.6	986	1876
Lysobacter	genus	0.004	0.36	164.9	657	236
Porphyromonas	genus	0.013	0.54	164.7	2156	1174
Psychrobacter glacialis	species	0.002	0.37	158	664	247

Summary

A large number of bacterial taxa exhibit shifts with P < 0.01 in association with this condition. The subsequent challenge is determining how to modulate these taxa, since the volume of candidates exceeds what most individuals can practically consider. Moreover, for many of the taxa identified, there is no published evidence in the U.S. National Library of Medicine describing how to alter their abundance.

A deep optimization model, such as the one implemented on the Microbiome Taxa R2 site, can be used to inform probiotic selection. This model provides coverage for each identified taxon and infers which probiotics are most likely to shift their levels. Its output may then be integrated with more conventional recommendations derived from literature indexed in the U.S. National Library of Medicine where such evidence exists, with the two recommendation sets reconciled by giving priority to probiotic-based suggestions.

Development of a dedicated database based on Biomesight samples is in progress. The current model uses data contributed by PrecisionBiome, and datasets generated with differing laboratory processing pipelines cannot be safely combined, as discussed in The taxonomy nightmare before Christmas…. Once the Biomesight-specific database is complete, an option for generating (offline-only) personalized suggestions will be added to the Microbiome Prescription website.

Probiotics Suggestions

The following are based on a simplified algorithm using R2 data for Biomesight. These are tentative numbers subject to future refinements. Bacteria listed are only for probiotics detected with Biomesight tests. Probiotics include some that are available only in some countries and some that are pending approval for retail sale.

• Good Count: Number of bacteria expected to shift in desired direction
• Bad Count: Number of bacteria expected to shift in wrong direction
• Impact: Estimator of impact based on Chi-2, Slope and R2 vectors

Probiotic Species	Impact	Good Count	Bad Count
Faecalibacterium prausnitzii	434.71	8	0
Bifidobacterium breve	64.25	17	0
Bifidobacterium longum	57.5	20	0
Bifidobacterium adolescentis	42.25	17	0
Bifidobacterium bifidum	13.35	12	2
Bifidobacterium catenulatum	11.39	11	0
Bifidobacterium animalis	6.14	7	0
Enterococcus faecium	2.64	23	25
Escherichia coli	1.67	5	8
Enterococcus faecalis	1.37	51	44
Limosilactobacillus reuteri	0.81	16	17
Streptococcus thermophilus	0.73	4	0
Bacillus subtilis	0.52	28	32
Veillonella atypica	0.5	9	2
Lactobacillus helveticus	0.34	46	44
Enterococcus durans	0.3	25	21
Lactococcus lactis	0.22	6	1
Heyndrickxia coagulans	0.12	10	11
Lacticaseibacillus paracasei	-0.02	3	6
Lactiplantibacillus pentosus	-0.02	2	2
Lactiplantibacillus plantarum	-0.06	3	5
Bifidobacterium pseudocatenulatum	-0.08	19	13
Lacticaseibacillus rhamnosus	-0.09	2	4
Lacticaseibacillus casei	-0.13	3	7
Leuconostoc mesenteroides	-0.24	4	11
Ligilactobacillus salivarius	-0.32	1	9
Clostridium butyricum	-0.4	15	20
Odoribacter laneus	-0.55	0	3
Lactobacillus crispatus	-0.65	7	24
Lactobacillus acidophilus	-1.02	9	29
Segatella copri	-1.1	4	2
Limosilactobacillus vaginalis	-1.19	25	45
Bacteroides thetaiotaomicron	-2.21	0	5
Bacteroides uniformis	-2.77	0	6
Lactobacillus jensenii	-3.6	22	51
Pediococcus acidilactici	-19.95	24	40
Blautia wexlerae	-19.99	3	2
Parabacteroides goldsteinii	-22.17	1	15
Akkermansia muciniphila	-27.03	2	26
Parabacteroides distasonis	-35.63	0	10
Blautia hansenii	-39.03	4	10
Lactobacillus johnsonii	-41.55	31	41

This document presents the results of statistical analysis on symptoms from viable, self-annotated Biomesight microbiome samples. The methodology for data acquisition is outlined in New Standards for Microbiome Analysis?.​

Tables have been refined to display only genus- and species-level taxa, the 20 most prominent entries per group, and associations achieving statistical significance (P < 0.01).​

The following sections provide the processed data, accompanied by guidance on interpretation and application. Counts of significant bacterial taxa are included, reflecting the application of non-standard but rigorously validated statistical approaches to extensive sample and reference populations, where statistical power derives from dataset scale.

Significance	Genus
p < 0.01	226
p < 0.001	199
p < 0.0001	176
p < 0.00001	157

Averages and Medians

I prefer medians over averages. Medians are the values where half of the people have less and half has more. If the data was a bell-curve, then the values will almost be the same… with bacteria that happens rarely. Look at Bacteroides below, we see that the average is above and the median below.

If symptom median is higher than reference median, it means there is more of this bacteria. If lower, then less. This ignores how often the bacteria is seen (we average only over reports).

tax_name	Rank	Symptom Avarage	Reference Average	Symptom Median	Reference Median
Bacteroides	genus	28.633	25.842	24.114	27.576
Phocaeicola vulgatus	species	6.593	5.744	3.347	4.644
Phocaeicola	genus	11.777	10.78	9.291	10.397
Bacteroides uniformis	species	3.031	2.707	1.52	2.104
Parabacteroides	genus	2.552	2.631	1.714	2.058
Bacteroides thetaiotaomicron	species	1.27	1.056	0.455	0.678
Clostridium	genus	2.004	1.847	1.352	1.569
Oscillospira	genus	2.45	2.348	1.944	2.144
Bacteroides cellulosilyticus	species	1.032	0.84	0.07	0.237
Parabacteroides merdae	species	0.854	0.737	0.287	0.44
Coprococcus	genus	1.071	1.461	0.739	0.59
Pedobacter	genus	1.232	0.98	0.548	0.695
Novispirillum	genus	0.816	0.872	0.087	0.229
Insolitispirillum	genus	0.816	0.873	0.089	0.229
Insolitispirillum peregrinum	species	0.816	0.873	0.089	0.229
Bacteroides caccae	species	1.135	0.854	0.281	0.38
Bifidobacterium	genus	0.513	0.969	0.134	0.048
Bilophila	genus	0.395	0.347	0.206	0.276
Parabacteroides goldsteinii	species	0.565	0.571	0.13	0.194
Bilophila wadsworthia	species	0.376	0.339	0.197	0.256

Bacteria Incidence – How often is it reported

The common sense belief is that if a bacteria is reported more often, then the amount should be higher. This is often not true. The microbiome is a complex thing. Look at Bacteroides uniformis below, we see that the average is above and the median below

tax_name	Rank	Incidence Odds Ratio	Chi2	Symptoms %	Reference %
Bifidobacterium catenulatum	species	0.69	8.3	25.2	36.2
Thiomonas thermosulfata	species	1.41	8	29.4	20.9
Aggregatibacter	genus	0.65	7.8	16	24.4
Desulfurispirillum	genus	1.45	8.4	25.5	17.5
Desulfurispirillum alkaliphilum	species	1.45	8.1	25.2	17.4
Actinobacillus pleuropneumoniae	species	0.6	8.4	11	18.5
Bifidobacterium cuniculi	species	0.62	7.8	12	19.3
Desulfonatronovibrio	genus	1.48	7.5	19.9	13.5
Erysipelothrix inopinata	species	1.47	7.4	19.9	13.5
Paraprevotella xylaniphila	species	1.6	9.2	16.6	10.4
Candidatus Phytoplasma phoenicium	species	1.59	8.7	16	10.1
Trichodesmium	genus	1.56	6.6	12.9	8.3

More or Less often based on Symptom Median All Incidence

This is a little more complex to understand. If we compute the mid point for people with the symptom, then if the bacteria was not involved then half of the reference should be above this value and half below this value. If not, it means that the symptom tends to over or under growth.

tax_name	Rank	Symptom Median	Odds Ratio	Chi2	Below	Above
Alcanivorax	genus	0.002	0.26	75.2	375	98
Isoalcanivorax	genus	0.002	0.26	74.5	365	95
Isoalcanivorax indicus	species	0.002	0.26	74.5	365	95
Nostoc flagelliforme	species	0.002	0.27	67.3	316	84
Salidesulfovibrio	genus	0.002	0.3	63.2	381	115
Salidesulfovibrio brasiliensis	species	0.002	0.3	63.2	381	115
Niabella aurantiaca	species	0.002	0.34	62.1	524	176
Mycoplasmopsis lipophila	species	0.002	0.27	61.2	277	75
Psychroflexus	genus	0.002	0.3	60.4	348	105
Psychroflexus gondwanensis	species	0.002	0.3	60.4	348	105
Deferribacter autotrophicus	species	0.002	0.31	59.6	374	117
Pelagicoccus croceus	species	0.002	0.32	59.2	380	120
Deferribacter	genus	0.002	0.32	59	377	119
Psychrobacter glacialis	species	0.002	0.37	55.3	636	238
Thermodesulfatator atlanticus	species	0.002	0.3	53.6	276	83
Thermodesulfatator	genus	0.002	0.3	53.6	276	83
Segetibacter aerophilus	species	0.002	0.34	53.1	364	122
Bacillus ferrariarum	species	0.002	0.34	52.1	356	120
Niabella	genus	0.002	0.38	51.8	560	212
Rickettsia marmionii Stenos et al. 2005	species	0.002	0.35	51.6	379	131

More or Less often based on Reference Median All Incidence

This is like the above, but with a different line in the sand. Instead of the median of those with the condition, we use the median of the reference set.

tax_name	Rank	Reference Median	Odds Ratio	Chi2	Below	Above
Psychrobacter glacialis	species	0.002	0.37	150	636	238
Niabella aurantiaca	species	0.002	0.34	148.2	524	176
Desulfotomaculum	genus	0.004	0.51	147.5	1394	711
Alcanivorax	genus	0.002	0.26	145.8	375	98
Isoalcanivorax	genus	0.002	0.26	142.8	365	95
Isoalcanivorax indicus	species	0.002	0.26	142.8	365	95
Niabella	genus	0.002	0.38	132.4	560	212
Salidesulfovibrio	genus	0.002	0.3	127.5	381	115
Salidesulfovibrio brasiliensis	species	0.002	0.3	127.5	381	115
Bacteroides cellulosilyticus	species	0.237	0.59	127.2	2243	1325
Actinopolyspora	genus	0.002	0.38	125.5	524	198
Bifidobacterium	genus	0.048	1.68	125.3	1392	2332
Nostoc flagelliforme	species	0.002	0.27	122.8	316	84
Geobacillus	genus	0.003	0.43	122.4	672	291
Bacteroides heparinolyticus	species	0.003	0.49	122.1	928	454
Erysipelothrix muris	species	0.014	0.59	121.6	2029	1195
Pelagicoccus croceus	species	0.002	0.32	120.7	380	120
Deferribacter autotrophicus	species	0.002	0.31	120.3	374	117
Deferribacter	genus	0.002	0.32	119.9	377	119
Psychroflexus	genus	0.002	0.3	117.5	348	105

More or Less often based on Symptom Median High Incidence

Above we see that many of the top bacteria identified are sparse, that is not reported often. We then restrict them to those that occur above 50% or the time.

tax_name	Rank	Symptom Median Freq	Odds Ratio	Chi2	Below	Above
Clostridium taeniosporum	species	0.003	0.62	16	1295	805
Tetragenococcus doogicus	species	0.003	0.67	11.5	1316	881
Dethiosulfovibrio	genus	0.004	0.67	11.3	1457	981
Hydrocarboniphaga daqingensis	species	0.004	0.71	8.7	1549	1096
Mycoplasmopsis	genus	0.005	0.72	8	1709	1230

More or Less often based on Reference Median High Incidence

Above we see that many of the top bacteria identified are sparse, that is not reported often. We then restrict them to those that occur above 50% or the time.

tax_name	Rank	Reference Median Freq	Odds Ratio	Chi2	Below	Above
Psychrobacter glacialis	species	0.002	0.37	150	636	238
Niabella aurantiaca	species	0.002	0.34	148.2	524	176
Desulfotomaculum	genus	0.004	0.51	147.5	1394	711
Alcanivorax	genus	0.002	0.26	145.8	375	98
Isoalcanivorax	genus	0.002	0.26	142.8	365	95
Isoalcanivorax indicus	species	0.002	0.26	142.8	365	95
Niabella	genus	0.002	0.38	132.4	560	212
Salidesulfovibrio	genus	0.002	0.3	127.5	381	115
Salidesulfovibrio brasiliensis	species	0.002	0.3	127.5	381	115
Bacteroides cellulosilyticus	species	0.237	0.59	127.2	2243	1325
Actinopolyspora	genus	0.002	0.38	125.5	524	198
Bifidobacterium	genus	0.048	1.68	125.3	1392	2332
Nostoc flagelliforme	species	0.002	0.27	122.8	316	84
Geobacillus	genus	0.003	0.43	122.4	672	291
Bacteroides heparinolyticus	species	0.003	0.49	122.1	928	454
Erysipelothrix muris	species	0.014	0.59	121.6	2029	1195
Pelagicoccus croceus	species	0.002	0.32	120.7	380	120
Deferribacter autotrophicus	species	0.002	0.31	120.3	374	117
Deferribacter	genus	0.002	0.32	119.9	377	119
Psychroflexus	genus	0.002	0.3	117.5	348	105

Summary

A large number of bacterial taxa exhibit shifts with P < 0.01 in association with this condition. The subsequent challenge is determining how to modulate these taxa, since the volume of candidates exceeds what most individuals can practically consider. Moreover, for many of the taxa identified, there is no published evidence in the U.S. National Library of Medicine describing how to alter their abundance.

A deep optimization model, such as the one implemented on the Microbiome Taxa R2 site, can be used to inform probiotic selection. This model provides coverage for each identified taxon and infers which probiotics are most likely to shift their levels. Its output may then be integrated with more conventional recommendations derived from literature indexed in the U.S. National Library of Medicine where such evidence exists, with the two recommendation sets reconciled by giving priority to probiotic-based suggestions.

Development of a dedicated database based on Biomesight samples is in progress. The current model uses data contributed by PrecisionBiome, and datasets generated with differing laboratory processing pipelines cannot be safely combined, as discussed in The taxonomy nightmare before Christmas…. Once the Biomesight-specific database is complete, an option for generating (offline-only) personalized suggestions will be added to the Microbiome Prescription website.

Probiotics Suggestions

The following are based on a simplified algorithm using R2 data for Biomesight. These are tentative numbers subject to future refinements. Bacteria listed are only for probiotics detected with Biomesight tests. Probiotics include some that are available only in some countries and some that are pending approval for retail sale.

• Good Count: Number of bacteria expected to shift in desired direction
• Bad Count: Number of bacteria expected to shift in wrong direction
• Impact: Estimator of impact based on Chi-2, Slope and R2 vectors

Probiotic Species	Impact	Good Count	Bad Count
Blautia hansenii	92.34	11	1
Bifidobacterium breve	70.82	17	3
Faecalibacterium prausnitzii	67.96	12	0
Bifidobacterium longum	64.14	19	0
Blautia wexlerae	50.13	6	0
Bifidobacterium adolescentis	46.86	15	0
Segatella copri	39.48	7	0
Enterococcus faecalis	30.68	60	31
Bifidobacterium bifidum	14.43	15	3
Lactobacillus helveticus	13.52	47	77
Bifidobacterium catenulatum	13.5	13	2
Escherichia coli	12.84	13	0
Enterococcus faecium	7.35	20	26
Bifidobacterium animalis	6.59	8	1
Lactobacillus johnsonii	1.25	40	46
Streptococcus thermophilus	1.17	3	0
Veillonella atypica	0.93	9	0
Bacillus subtilis	0.85	33	46
Clostridium butyricum	0.74	24	25
Enterococcus durans	0.45	36	23
Ligilactobacillus salivarius	0.3	12	11
Leuconostoc mesenteroides	0.23	17	12
Lactococcus lactis	0.13	5	11
Limosilactobacillus fermentum	0.08	10	12
Lacticaseibacillus paracasei	-0.08	12	28
Lacticaseibacillus casei	-0.1	0	9
Heyndrickxia coagulans	-0.22	20	35
Lactiplantibacillus plantarum	-0.26	2	9
Bifidobacterium pseudocatenulatum	-0.41	18	35
Lactiplantibacillus pentosus	-0.44	8	22
Lactobacillus crispatus	-0.46	1	32
Lactobacillus acidophilus	-0.52	11	28
Lacticaseibacillus rhamnosus	-0.6	3	28
Lactobacillus jensenii	-1.44	19	41
Odoribacter laneus	-1.54	0	4
Limosilactobacillus vaginalis	-1.79	30	52
Limosilactobacillus reuteri	-2.22	31	46
Pediococcus acidilactici	-6.8	25	23
Akkermansia muciniphila	-13.33	2	32
Parabacteroides goldsteinii	-41.6	0	18
Parabacteroides distasonis	-60.25	0	7
Bacteroides uniformis	-250.42	0	11
Bacteroides thetaiotaomicron	-261.55	1	11

This document presents the results of statistical analysis on symptoms from viable, self-annotated Biomesight microbiome samples. The methodology for data acquisition is outlined in New Standards for Microbiome Analysis?.​

Tables have been refined to display only genus- and species-level taxa, the 20 most prominent entries per group, and associations achieving statistical significance (P < 0.01).​

The following sections provide the processed data, accompanied by guidance on interpretation and application. Counts of significant bacterial taxa are included, reflecting the application of non-standard but rigorously validated statistical approaches to extensive sample and reference populations, where statistical power derives from dataset scale.

Significance	Genus
p < 0.01	202
p < 0.001	173
p < 0.0001	145
p < 0.00001	129

Averages and Medians

I prefer medians over averages. Medians are the values where half of the people have less and half has more. If the data was a bell-curve, then the values will almost be the same… with bacteria that happens rarely. Look at Bacteroides uniformis below, we see that the average is above and the median below.

If symptom median is higher than reference median, it means there is more of this bacteria. If lower, then less. This ignores how often the bacteria is seen (we average only over reports).

tax_name	Rank	Symptom Avarage	Reference Average	Symptom Median	Reference Median
Bacteroides	genus	27.633	25.884	24.121	26.729
Bacteroides uniformis	species	2.985	2.705	1.527	2.059
Novispirillum	genus	0.784	0.876	0.085	0.174
Insolitispirillum	genus	0.784	0.877	0.086	0.174
Insolitispirillum peregrinum	species	0.784	0.877	0.086	0.174
Bacteroides cellulosilyticus	species	1.024	0.836	0.073	0.151
Bilophila	genus	0.376	0.348	0.206	0.265
Bifidobacterium	genus	0.607	0.969	0.131	0.072
Bilophila wadsworthia	species	0.365	0.339	0.196	0.255
Blautia obeum	species	0.654	0.563	0.228	0.281
Hathewaya histolytica	species	0.318	0.275	0.154	0.188
Hathewaya	genus	0.318	0.275	0.154	0.188
Lachnobacterium	genus	0.327	0.32	0.075	0.049
Anaerotruncus	genus	0.213	0.184	0.136	0.159
Bifidobacterium longum	species	0.248	0.33	0.05	0.029
Oribacterium	genus	0.123	0.131	0.074	0.053
Anaerotruncus colihominis	species	0.2	0.173	0.133	0.153
Oribacterium sinus	species	0.116	0.127	0.072	0.053
Odoribacter	genus	0.194	0.197	0.123	0.139
Megamonas	genus	0.42	0.439	0.003	0.018

Bacteria Incidence – How often is it reported

The common sense belief is that if a bacteria is reported more often, then the amount should be higher. This is often not true. The microbiome is a complex thing. Look at Bacteroides uniformis below, we see that the average is above and the median below

tax_name	Rank	Incidence Odds Ratio	Chi2	Symptoms %	Reference %
Prevotella bivia	species	1.42	9.5	26.9	19

More or Less often based on Symptom Median All Incidence

This is a little more complex to understand. If we compute the mid point for people with the symptom, then if the bacteria was not involved then half of the reference should be above this value and half below this value. If not, it means that the symptom tends to over or under growth.

tax_name	Rank	Symptom Median	Odds Ratio	Chi2	Below	Above
Alcanivorax	genus	0.002	0.27	77.8	360	99
Isoalcanivorax	genus	0.002	0.27	77.3	349	95
Isoalcanivorax indicus	species	0.002	0.27	77.3	349	95
Niabella aurantiaca	species	0.002	0.34	69.1	505	172
Psychroflexus	genus	0.002	0.3	66.9	345	105
Psychroflexus gondwanensis	species	0.002	0.3	66.9	345	105
Salidesulfovibrio	genus	0.002	0.32	65.7	376	120
Salidesulfovibrio brasiliensis	species	0.002	0.32	65.7	376	120
Deferribacter autotrophicus	species	0.002	0.32	65.4	366	116
Deferribacter	genus	0.002	0.32	64.5	368	118
Psychrobacter glacialis	species	0.002	0.38	62.5	623	237
Pelagicoccus croceus	species	0.002	0.32	62.3	357	116
Rickettsia marmionii Stenos et al. 2005	species	0.002	0.34	60.6	372	125
Bacillus ferrariarum	species	0.002	0.34	58.4	350	118
Niabella	genus	0.002	0.39	56.6	542	211
Actinopolyspora	genus	0.002	0.39	55.8	505	195
Chromatium	genus	0.002	0.39	54.7	480	185
Chromatium weissei	species	0.002	0.39	54.4	479	185
Viridibacillus neidei	species	0.002	0.38	53.7	445	170
Segetibacter aerophilus	species	0.002	0.36	53.7	353	126

More or Less often based on Reference Median All Incidence

This is like the above, but with a different line in the sand. Instead of the median of those with the condition, we use the median of the reference set.

tax_name	Rank	Reference Median	Odds Ratio	Chi2	Below	Above
Methylobacillus glycogenes	species	0.003	0.4	217.9	1190	477
Methylobacillus	genus	0.003	0.42	203.2	1190	496
Psychrobacter glacialis	species	0.002	0.38	143.2	623	237
Niabella aurantiaca	species	0.002	0.34	140.5	505	172
Alcanivorax	genus	0.002	0.27	133.4	360	99
Isoalcanivorax	genus	0.002	0.27	131	349	95
Isoalcanivorax indicus	species	0.002	0.27	131	349	95
Niabella	genus	0.002	0.39	122.8	542	211
Salidesulfovibrio	genus	0.002	0.32	117.8	376	120
Salidesulfovibrio brasiliensis	species	0.002	0.32	117.8	376	120
Actinopolyspora	genus	0.002	0.39	117.1	505	195
Deferribacter autotrophicus	species	0.002	0.32	115.9	366	116
Psychroflexus	genus	0.002	0.3	115.2	345	105
Psychroflexus gondwanensis	species	0.002	0.3	115.2	345	105
Deferribacter	genus	0.002	0.32	114.9	368	118
Chromatium	genus	0.002	0.39	112.4	480	185
Chromatium weissei	species	0.002	0.39	111.9	479	185
Helicobacter suncus	species	0.002	0.46	111.6	717	333
Pelagicoccus croceus	species	0.002	0.32	110	357	116
Rickettsia marmionii Stenos et al. 2005	species	0.002	0.34	109.4	372	125

More or Less often based on Symptom Median High Incidence

Above we see that many of the top bacteria identified are sparse, that is not reported often. We then restrict them to those that occur above 50% or the time.

tax_name	Rank	Symptom Median Freq	Odds Ratio	Chi2	Below	Above
Clostridium taeniosporum	species	0.003	0.61	21	1269	777
Dethiosulfovibrio	genus	0.004	0.67	14.6	1417	946
Tetragenococcus doogicus	species	0.003	0.68	12.6	1279	876
Hydrocarboniphaga daqingensis	species	0.004	0.72	9.8	1499	1078
Mycoplasmopsis	genus	0.005	0.72	9.7	1661	1199
Pediococcus	genus	0.004	0.75	7.2	1217	913
Propionispora hippei	species	0.005	0.76	6.8	1449	1101
Propionispora	genus	0.005	0.76	6.7	1448	1102

More or Less often based on Reference Median High Incidence

Above we see that many of the top bacteria identified are sparse, that is not reported often. We then restrict them to those that occur above 50% or the time.

Methylobacillus glycogenes	species	0.003	0.4	217.9	1190	477
Methylobacillus	genus	0.003	0.42	203.2	1190	496
Psychrobacter glacialis	species	0.002	0.38	143.2	623	237
Niabella aurantiaca	species	0.002	0.34	140.5	505	172
Alcanivorax	genus	0.002	0.27	133.4	360	99
Isoalcanivorax	genus	0.002	0.27	131	349	95
Isoalcanivorax indicus	species	0.002	0.27	131	349	95
Niabella	genus	0.002	0.39	122.8	542	211
Salidesulfovibrio	genus	0.002	0.32	117.8	376	120
Salidesulfovibrio brasiliensis	species	0.002	0.32	117.8	376	120
Actinopolyspora	genus	0.002	0.39	117.1	505	195
Deferribacter autotrophicus	species	0.002	0.32	115.9	366	116
Psychroflexus	genus	0.002	0.3	115.2	345	105
Psychroflexus gondwanensis	species	0.002	0.3	115.2	345	105
Deferribacter	genus	0.002	0.32	114.9	368	118
Chromatium	genus	0.002	0.39	112.4	480	185
Chromatium weissei	species	0.002	0.39	111.9	479	185
Helicobacter suncus	species	0.002	0.46	111.6	717	333
Pelagicoccus croceus	species	0.002	0.32	110	357	116
Rickettsia marmionii Stenos et al. 2005	species	0.002	0.34	109.4	372	125

Summary

A large number of bacterial taxa exhibit shifts with P < 0.01 in association with this condition. The subsequent challenge is determining how to modulate these taxa, since the volume of candidates exceeds what most individuals can practically consider. Moreover, for many of the taxa identified, there is no published evidence in the U.S. National Library of Medicine describing how to alter their abundance.

A deep optimization model, such as the one implemented on the Microbiome Taxa R2 site, can be used to inform probiotic selection. This model provides coverage for each identified taxon and infers which probiotics are most likely to shift their levels. Its output may then be integrated with more conventional recommendations derived from literature indexed in the U.S. National Library of Medicine where such evidence exists, with the two recommendation sets reconciled by giving priority to probiotic-based suggestions.

Development of a dedicated database based on Biomesight samples is in progress. The current model uses data contributed by PrecisionBiome, and datasets generated with differing laboratory processing pipelines cannot be safely combined, as discussed in The taxonomy nightmare before Christmas…. Once the Biomesight-specific database is complete, an option for generating (offline-only) personalized suggestions will be added to the Microbiome Prescription website.

Probiotics Suggestions

The following are based on a simplified algorithm using R2 data for Biomesight. These are tentative numbers subject to future refinements. Bacteria listed are only for probiotics detected with Biomesight tests. Probiotics include some that are available only in some countries and some that are pending approval for retail sale.

• Good Count: Number of bacteria expected to shift in desired direction
• Bad Count: Number of bacteria expected to shift in wrong direction
• Impact: Estimator of impact based on Chi-2, Slope and R2 vectors

Probiotic Species	Impact	Good Count	Bad Count
Bifidobacterium breve	38.64	13	0
Bifidobacterium longum	34.39	12	0
Bifidobacterium adolescentis	25.49	11	0
Segatella copri	21.76	5	0
Akkermansia muciniphila	16.19	13	7
Lactobacillus helveticus	10.85	70	36
Bifidobacterium bifidum	7.8	16	2
Bifidobacterium catenulatum	6.98	14	0
Escherichia coli	3.79	8	0
Lactobacillus johnsonii	3.74	52	16
Bifidobacterium animalis	3.53	8	0
Pediococcus acidilactici	3.33	31	26
Enterococcus faecalis	2.72	47	26
Enterococcus durans	2.31	50	14
Enterococcus faecium	1.83	28	22
Streptococcus thermophilus	1.28	2	1
Clostridium butyricum	0.9	24	11
Limosilactobacillus reuteri	0.82	27	12
Lactococcus lactis	0.28	6	3
Leuconostoc mesenteroides	0.26	18	12
Lacticaseibacillus paracasei	0.25	20	8
Ligilactobacillus salivarius	0.25	16	5
Limosilactobacillus fermentum	0.23	19	14
Bacillus subtilis	0.15	37	29
Bifidobacterium pseudocatenulatum	0.14	29	9
Lactiplantibacillus plantarum	0.12	3	1
Limosilactobacillus vaginalis	0.11	43	36
Lactiplantibacillus pentosus	0.1	7	2
Veillonella atypica	0.09	3	1
Lacticaseibacillus casei	0.05	11	10
Lactobacillus crispatus	-0.02	14	15
Lactobacillus acidophilus	-0.08	18	13
Lacticaseibacillus rhamnosus	-0.1	2	8
Heyndrickxia coagulans	-0.16	14	21
Odoribacter laneus	-0.4	0	2
Parabacteroides goldsteinii	-0.43	2	4
Lactobacillus jensenii	-1.69	36	30
Faecalibacterium prausnitzii	-3.39	1	4
Blautia hansenii	-12.12	2	2
Blautia wexlerae	-12.46	0	1
Bacteroides uniformis	-136.48	0	9
Bacteroides thetaiotaomicron	-148.97	0	9

This document presents the results of statistical analysis on symptoms from viable, self-annotated Biomesight microbiome samples. The methodology for data acquisition is outlined in New Standards for Microbiome Analysis?.​

Tables have been refined to display only genus- and species-level taxa, the 20 most prominent entries per group, and associations achieving statistical significance (P < 0.01).​

The following sections provide the processed data, accompanied by guidance on interpretation and application. Counts of significant bacterial taxa are included, reflecting the application of non-standard but rigorously validated statistical approaches to extensive sample and reference populations, where statistical power derives from dataset scale.

Significance	Genus
p < 0.01	182
p < 0.001	164
p < 0.0001	146
p < 0.00001	130

Averages and Medians

I prefer medians over averages. Medians are the values where half of the people have less and half has more. If the data was a bell-curve, then the values will almost be the same… with bacteria that happens rarely. Look at the bacterua below, we see that for some the average is above and the median below. Should one increase or decrease this bacteria?

If symptom median is higher than reference median, it means there is more of this bacteria. If lower, then less. This ignores how often the bacteria is seen (we average only over reports). IMHO using average value instead of median will often result in a worse situation for the patient

tax_name	Rank	Symptom Avarage	Reference Average	Symptom Median	Reference Median
Bacteroides	genus	31.102	25.834	24.217	30.409
Faecalibacterium	genus	10.531	12.875	12.155	9.178
Faecalibacterium prausnitzii	species	10.193	12.301	11.47	8.958
Phocaeicola vulgatus	species	6.283	5.788	3.427	4.247
Bacteroides uniformis	species	3.229	2.71	1.559	2.11
Ruminococcus	genus	5.96	5.586	4.394	3.874
Coprococcus	genus	1.313	1.436	0.738	0.483
Clostridium	genus	2.067	1.85	1.359	1.612
Phocaeicola dorei	species	3.715	2.873	0.412	0.196
Bacteroides thetaiotaomicron	species	1.612	1.049	0.463	0.593
Bacteroides cellulosilyticus	species	1.092	0.844	0.075	0.179
Lachnospira pectinoschiza	species	0.549	0.668	0.337	0.245
Ruminococcus bromii	species	0.784	0.791	0.174	0.083
Bifidobacterium	genus	0.574	0.95	0.128	0.045
Bilophila wadsworthia	species	0.387	0.34	0.199	0.273
Bilophila	genus	0.395	0.348	0.209	0.281
Lachnobacterium	genus	0.222	0.325	0.076	0.028
Sutterella wadsworthensis	species	0.711	0.657	0.059	0.011
Dorea	genus	0.494	0.482	0.292	0.336
Hathewaya	genus	0.381	0.275	0.155	0.198

Bacteria Incidence – How often is it reported

The common sense belief is that if a bacteria is reported more often, then the amount should be higher. This is often not true. The microbiome is a complex thing.

tax_name	Rank	Incidence Odds Ratio	Chi2	Symptoms %	Reference %
Halanaerobium	genus	1.58	7.7	26.4	16.7

More or Less often based on Symptom Median All Incidence

This is a little more complex to understand. If we compute the mid point for people with the symptom, then if the bacteria was not involved then half of the reference should be above this value and half below this value. If not, it means that the symptom tends to over or under growth.

tax_name	Rank	Symptom Median	Odds Ratio	Chi2	Below	Above
Isoalcanivorax	genus	0.002	0.26	53.9	371	98
Isoalcanivorax indicus	species	0.002	0.26	53.9	371	98
Alcanivorax	genus	0.002	0.27	53.8	382	102
Niabella aurantiaca	species	0.002	0.33	43.2	545	182
Pelagicoccus croceus	species	0.002	0.32	40.7	378	122
Psychrobacter glacialis	species	0.002	0.38	35.8	660	250
Niabella	genus	0.002	0.38	35	585	221
Viridibacillus neidei	species	0.002	0.38	32.9	472	179
Chromatium	genus	0.002	0.39	32.1	515	200
Chromatium weissei	species	0.002	0.39	32	514	200
Sporosarcina pasteurii	species	0.002	0.4	28.6	444	179
Thiorhodococcus	genus	0.002	0.42	27.4	578	245
Syntrophomonas sapovorans	species	0.002	0.42	27	536	227
Sporosarcina	genus	0.002	0.42	26.9	448	186
Lysinibacillus	genus	0.002	0.42	25.9	401	167
Thermodesulfovibrio thiophilus	species	0.002	0.45	23.5	540	243
Oenococcus	genus	0.002	0.46	22.6	601	277
Thermodesulfovibrio	genus	0.002	0.47	22.1	625	292
Helicobacter suncus	species	0.002	0.48	21.7	761	363
Viridibacillus	genus	0.002	0.5	17.6	486	242

More or Less often based on Reference Median All Incidence

This is like the above, but with a different line in the sand. Instead of the median of those with the condition, we use the median of the reference set.

tax_name	Rank	Reference Median	Odds Ratio	Chi2	Below	Above
Lachnobacterium	genus	0.028	2.07	243	1197	2474
Bifidobacterium longum	species	0.0135	2.14	228.2	900	1929
Paenibacillus	genus	0.003	0.38	208.1	999	384
Erysipelothrix	genus	0.017	0.51	205.7	2244	1135
Anaerobranca zavarzinii	species	0.005	1.98	195.5	1046	2066
Anaerobranca	genus	0.005	1.98	195.5	1046	2066
Erysipelothrix muris	species	0.016	0.52	191.4	2194	1135
Slackia	genus	0.009	1.9	182.1	1161	2202
Legionella shakespearei	species	0.003	0.37	158.9	659	243
Bacteroides	genus	30.409	0.57	158.2	2511	1422
Faecalibacterium	genus	9.178	1.76	157.2	1423	2508
Niabella aurantiaca	species	0.002	0.33	155.3	545	182
Eubacterium callanderi	species	0.007	0.54	155.2	1882	1016
Holdemania	genus	0.027	0.56	153.4	2222	1244
Psychrobacter glacialis	species	0.002	0.38	152.7	660	250
Bifidobacterium	genus	0.045	1.73	146.5	1413	2447
Methylonatrum	genus	0.004	0.54	142.5	1617	870
Methylonatrum kenyense	species	0.004	0.54	142.5	1617	870
Amedibacillus dolichus	species	0.022	0.54	141.7	1678	912
Amedibacillus	genus	0.022	0.54	141.4	1677	912

More or Less often based on Symptom Median High Incidence

Above we see that many of the top bacteria identified are sparse, that is not reported often. We then restrict them to those that occur above 50% or the time.

tax_name	Rank	Symptom Median Freq	Odds Ratio	Chi2	Below	Above
Clostridium taeniosporum	species	0.003	0.61	10.2	1349	827
Dethiosulfovibrio	genus	0.004	0.67	6.7	1505	1012

More or Less often based on Reference Median High Incidence

Above we see that many of the top bacteria identified are sparse, that is not reported often. We then restrict them to those that occur above 50% or the time.

tax_name	Rank	Reference Median Freq	Odds Ratio	Chi2	Below	Above
Lachnobacterium	genus	0.028	2.07	243	1197	2474
Bifidobacterium longum	species	0.0135	2.14	228.2	900	1929
Paenibacillus	genus	0.003	0.38	208.1	999	384
Erysipelothrix	genus	0.017	0.51	205.7	2244	1135
Anaerobranca zavarzinii	species	0.005	1.98	195.5	1046	2066
Anaerobranca	genus	0.005	1.98	195.5	1046	2066
Erysipelothrix muris	species	0.016	0.52	191.4	2194	1135
Slackia	genus	0.009	1.9	182.1	1161	2202
Legionella shakespearei	species	0.003	0.37	158.9	659	243
Bacteroides	genus	30.409	0.57	158.2	2511	1422
Faecalibacterium	genus	9.178	1.76	157.2	1423	2508
Niabella aurantiaca	species	0.002	0.33	155.3	545	182
Eubacterium callanderi	species	0.007	0.54	155.2	1882	1016
Holdemania	genus	0.027	0.56	153.4	2222	1244
Psychrobacter glacialis	species	0.002	0.38	152.7	660	250
Bifidobacterium	genus	0.045	1.73	146.5	1413	2447
Methylonatrum	genus	0.004	0.54	142.5	1617	870
Methylonatrum kenyense	species	0.004	0.54	142.5	1617	870
Amedibacillus dolichus	species	0.022	0.54	141.7	1678	912
Amedibacillus	genus	0.022	0.54	141.4	1677	912

Summary

A large number of bacterial taxa exhibit shifts with P < 0.01 in association with this condition. The subsequent challenge is determining how to modulate these taxa, since the volume of candidates exceeds what most individuals can practically consider. Moreover, for many of the taxa identified, there is no published evidence in the U.S. National Library of Medicine describing how to alter their abundance.

A deep optimization model, such as the one implemented on the Microbiome Taxa R2 site, can be used to inform probiotic selection. This model provides coverage for each identified taxon and infers which probiotics are most likely to shift their levels. Its output may then be integrated with more conventional recommendations derived from literature indexed in the U.S. National Library of Medicine where such evidence exists, with the two recommendation sets reconciled by giving priority to probiotic-based suggestions.

Development of a dedicated database based on Biomesight samples is in progress. The current model uses data contributed by PrecisionBiome, and datasets generated with differing laboratory processing pipelines cannot be safely combined, as discussed in The taxonomy nightmare before Christmas…. Once the Biomesight-specific database is complete, an option for generating (offline-only) personalized suggestions will be added to the Microbiome Prescription website.

Probiotics Suggestions

The following are based on a simplified algorithm using R2 data for Biomesight. These are tentative numbers subject to future refinements. Bacteria listed are only for probiotics detected with Biomesight tests. Probiotics include some that are available only in some countries and some that are pending approval for retail sale.

• Good Count: Number of bacteria expected to shift in desired direction
• Bad Count: Number of bacteria expected to shift in wrong direction
• Impact: Estimator of impact based on Chi-2, Slope and R2 vectors

Probiotic Species	Impact	Good Count	Bad Count
Faecalibacterium prausnitzii	842.75	13	0
Blautia hansenii	196.62	20	1
Blautia wexlerae	129.19	8	0
Segatella copri	96.35	8	0
Bifidobacterium breve	64.71	16	0
Bifidobacterium longum	57.42	17	0
Bifidobacterium adolescentis	42.42	16	0
Lactobacillus helveticus	39.07	98	34
Akkermansia muciniphila	23.53	24	4
Bifidobacterium bifidum	13.29	11	2
Bifidobacterium catenulatum	11.26	8	2
Escherichia coli	8.67	7	3
Bifidobacterium animalis	6.08	7	1
Bacillus subtilis	1.26	44	38
Clostridium butyricum	0.96	19	17
Veillonella atypica	0.83	6	6
Enterococcus faecium	0.76	22	14
Enterococcus durans	0.66	26	20
Limosilactobacillus fermentum	0.51	22	3
Streptococcus thermophilus	0.32	2	3
Bifidobacterium pseudocatenulatum	0.22	20	20
Limosilactobacillus vaginalis	0.13	37	38
Lactiplantibacillus pentosus	0.13	6	1
Ligilactobacillus salivarius	0.04	7	3
Lactiplantibacillus plantarum	-0.01	3	5
Lactococcus lactis	-0.03	4	5
Lactobacillus crispatus	-0.06	1	9
Lacticaseibacillus paracasei	-0.1	5	11
Lacticaseibacillus rhamnosus	-0.17	3	6
Lactobacillus acidophilus	-0.18	10	11
Leuconostoc mesenteroides	-0.19	7	11
Lacticaseibacillus casei	-0.22	0	8
Limosilactobacillus reuteri	-0.35	17	20
Lactobacillus jensenii	-0.74	26	24
Parabacteroides goldsteinii	-6.86	4	4
Pediococcus acidilactici	-9.8	21	48
Enterococcus faecalis	-17.66	59	51
Parabacteroides distasonis	-22.69	0	4
Lactobacillus johnsonii	-58.41	48	39
Bacteroides uniformis	-547.16	0	13
Bacteroides thetaiotaomicron	-585.02	0	11

This document presents the results of statistical analysis on symptoms from viable, self-annotated Biomesight microbiome samples. The methodology for data acquisition is outlined in New Standards for Microbiome Analysis?.​

Tables have been refined to display only genus- and species-level taxa, the 20 most prominent entries per group, and associations achieving statistical significance (P < 0.01).​

The following sections provide the processed data, accompanied by guidance on interpretation and application. Counts of significant bacterial taxa are included, reflecting the application of non-standard but rigorously validated statistical approaches to extensive sample and reference populations, where statistical power derives from dataset scale.

Significance	Genus
p < 0.01	223
p < 0.001	199
p < 0.0001	181
p < 0.00001	164

Averages and Medians

I prefer medians over averages. Medians are the values where half of the people have less and half has more. If the data was a bell-curve, then the values will almost be the same… with bacteria that happens rarely. Look at the bacterua below, we see that for some the average is above and the median below. Should one increase or decrease this bacteria?

If symptom median is higher than reference median, it means there is more of this bacteria. If lower, then less. This ignores how often the bacteria is seen (we average only over reports). IMHO using average value instead of median will often result in a worse situation for the patient

tax_name	Rank	Symptom Average	Reference Average	Symptom Median	Reference Median
Faecalibacterium prausnitzii	species	13.575	12.086	11.275	12.554
Phocaeicola dorei	species	3.484	2.854	0.395	0.746
Roseburia	genus	2.432	2.876	1.812	1.484
Lachnospira	genus	2.424	2.755	1.901	1.631
Roseburia faecis	species	0.855	1.239	0.594	0.378
Sutterella wadsworthensis	species	0.75	0.65	0.049	0.239
Coprococcus	genus	1.083	1.463	0.741	0.609
Pedobacter	genus	1.299	0.971	0.551	0.651
Blautia wexlerae	species	0.474	0.589	0.324	0.27
Anaeroplasma	genus	1.197	0.432	0.003	0.05
Dorea	genus	0.454	0.486	0.299	0.256
Parabacteroides goldsteinii	species	0.585	0.569	0.133	0.171
Thermicanus	genus	0.206	0.188	0.101	0.127
Odoribacter	genus	0.28	0.189	0.122	0.146
Bacteroides stercorirosoris	species	0.166	0.196	0.139	0.116
Collinsella aerofaciens	species	0.154	0.172	0.05	0.071
Acetivibrio alkalicellulosi	species	0.237	0.261	0.1	0.08
Acetivibrio	genus	0.246	0.27	0.105	0.085
Dorea formicigenerans	species	0.111	0.136	0.086	0.067
Anaerofilum	genus	0.23	0.273	0.109	0.092

Bacteria Incidence – How often is it reported

The common sense belief is that if a bacteria is reported more often, then the amount should be higher. This is often not true. The microbiome is a complex thing.

tax_name	Rank	Incidence Odds Ratio	Chi2	Symptoms %	Reference %
Lactococcus	genus	1.27	7.2	60.5	47.6
Sporotomaculum	genus	0.72	8.8	31.6	43.8
Sporotomaculum syntrophicum	species	0.73	8.5	31.3	43.2
Enterobacter hormaechei	species	0.7	6.9	18	25.8
Actinopolyspora	genus	0.58	9.9	10.6	18.2
Rothia mucilaginosa	species	0.62	8.3	11.7	18.8
Citrobacter	genus	0.65	7.1	12.5	19.2
Peptoniphilus lacrimalis	species	1.47	8.2	20.7	14.1
Chromatium	genus	0.63	7.3	11.2	17.6
Chromatium weissei	species	0.64	7.3	11.2	17.6
Anaerococcus hydrogenalis	species	1.48	6.9	16.1	10.9

More or Less often based on Symptom Median All Incidence

This is a little more complex to understand. If we compute the mid point for people with the symptom, then if the bacteria was not involved then half of the reference should be above this value and half below this value. If not, it means that the symptom tends to over or under growth.

tax_name	Rank	Symptom Median	Odds Ratio	Chi2	Below	Above
Isoalcanivorax	genus	0.002	0.27	76.7	370	99
Isoalcanivorax indicus	species	0.002	0.27	76.7	370	99
Alcanivorax	genus	0.002	0.27	76.6	380	103
Nostoc flagelliforme	species	0.002	0.25	74.8	311	78
Niabella aurantiaca	species	0.002	0.32	71.5	534	172
Pelagicoccus croceus	species	0.002	0.3	66	378	115
Psychrobacter glacialis	species	0.002	0.36	65.8	654	233
Deferribacter autotrophicus	species	0.002	0.31	65.2	378	116
Deferribacter	genus	0.002	0.31	64.6	381	118
Salidesulfovibrio	genus	0.002	0.32	63.2	386	122
Salidesulfovibrio brasiliensis	species	0.002	0.32	63.2	386	122
Actinopolyspora	genus	0.002	0.36	60.6	537	192
Niabella	genus	0.002	0.36	60.1	571	208
Rickettsia marmionii Stenos et al. 2005	species	0.002	0.33	60	394	130
Lentibacillus	genus	0.002	0.36	57.6	500	181
Psychroflexus	genus	0.002	0.32	57.5	343	111
Psychroflexus gondwanensis	species	0.002	0.32	57.5	343	111
Lentibacillus salinarum	species	0.002	0.36	57.2	485	175
Viridibacillus neidei	species	0.002	0.36	56.6	463	166
Bacillus ferrariarum	species	0.002	0.34	56.1	361	121

More or Less often based on Reference Median All Incidence

This is like the above, but with a different line in the sand. Instead of the median of those with the condition, we use the median of the reference set.

tax_name	Rank	Reference Median	Odds Ratio	Chi2	Below	Above
Actinopolyspora	genus	0.003	0.16	326.3	628	101
Nostoc	genus	0.003	0.33	281.5	1134	376
Flammeovirga	genus	0.003	0.35	186.1	742	261
Asticcacaulis	genus	0.003	0.42	185.9	1064	446
Flammeovirga pacifica	species	0.003	0.35	185.6	741	261
Planococcus	genus	0.003	0.32	182.4	613	194
Planococcus columbae	species	0.003	0.31	176.2	575	179
Streptococcus oralis	species	0.003	0.48	167.3	1358	652
Psychrobacter glacialis	species	0.002	0.36	164.8	654	233
Niabella aurantiaca	species	0.002	0.32	158.7	534	172
Clostridium tepidiprofundi	species	0.003	0.38	152.9	659	248
Niabella	genus	0.002	0.36	142.4	571	208
Alcanivorax	genus	0.002	0.27	142.3	380	103
Isoalcanivorax	genus	0.002	0.27	140.7	370	99
Isoalcanivorax indicus	species	0.002	0.27	140.7	370	99
Atopobium fossor	species	0.003	0.37	138.2	555	203
Lentibacillus	genus	0.002	0.36	128.3	500	181
Nostoc flagelliforme	species	0.002	0.25	127.5	311	78
Desulfitobacterium	genus	0.005	0.38	126.8	525	197
Lentibacillus salinarum	species	0.002	0.36	125.5	485	175

More or Less often based on Symptom Median High Incidence

Above we see that many of the top bacteria identified are sparse, that is not reported often. We then restrict them to those that occur above 50% or the time.

tax_name	Rank	Symptom Median Freq	Odds Ratio	Chi2	Below	Above
Clostridium taeniosporum	species	0.003	0.62	17.5	1284	800
Dethiosulfovibrio	genus	0.004	0.67	12.7	1433	961
Tetragenococcus doogicus	species	0.003	0.68	11.8	1289	875
Mycoplasmopsis	genus	0.005	0.7	10	1707	1203
Hydrocarboniphaga daqingensis	species	0.004	0.72	9	1533	1097
Tetragenococcus	genus	0.003	0.74	6.8	1270	946
Pediococcus	genus	0.004	0.75	6.6	1239	926

More or Less often based on Reference Median High Incidence

Above we see that many of the top bacteria identified are sparse, that is not reported often. We then restrict them to those that occur above 50% or the time.

tax_name	Rank	Reference Median Freq	Odds Ratio	Chi2	Below	Above
Actinopolyspora	genus	0.003	0.16	326.3	628	101
Nostoc	genus	0.003	0.33	281.5	1134	376
Flammeovirga	genus	0.003	0.35	186.1	742	261
Asticcacaulis	genus	0.003	0.42	185.9	1064	446
Flammeovirga pacifica	species	0.003	0.35	185.6	741	261
Planococcus	genus	0.003	0.32	182.4	613	194
Planococcus columbae	species	0.003	0.31	176.2	575	179
Streptococcus oralis	species	0.003	0.48	167.3	1358	652
Psychrobacter glacialis	species	0.002	0.36	164.8	654	233
Niabella aurantiaca	species	0.002	0.32	158.7	534	172
Clostridium tepidiprofundi	species	0.003	0.38	152.9	659	248
Niabella	genus	0.002	0.36	142.4	571	208
Alcanivorax	genus	0.002	0.27	142.3	380	103
Isoalcanivorax	genus	0.002	0.27	140.7	370	99
Isoalcanivorax indicus	species	0.002	0.27	140.7	370	99
Atopobium fossor	species	0.003	0.37	138.2	555	203
Lentibacillus	genus	0.002	0.36	128.3	500	181
Nostoc flagelliforme	species	0.002	0.25	127.5	311	78
Desulfitobacterium	genus	0.005	0.38	126.8	525	197
Lentibacillus salinarum	species	0.002	0.36	125.5	485	175

Summary

A large number of bacterial taxa exhibit shifts with P < 0.01 in association with this condition. The subsequent challenge is determining how to modulate these taxa, since the volume of candidates exceeds what most individuals can practically consider. Moreover, for many of the taxa identified, there is no published evidence in the U.S. National Library of Medicine describing how to alter their abundance.

A deep optimization model, such as the one implemented on the Microbiome Taxa R2 site, can be used to inform probiotic selection. This model provides coverage for each identified taxon and infers which probiotics are most likely to shift their levels. Its output may then be integrated with more conventional recommendations derived from literature indexed in the U.S. National Library of Medicine where such evidence exists, with the two recommendation sets reconciled by giving priority to probiotic-based suggestions.

Development of a dedicated database based on Biomesight samples is in progress. The current model uses data contributed by PrecisionBiome, and datasets generated with differing laboratory processing pipelines cannot be safely combined, as discussed in The taxonomy nightmare before Christmas…. Once the Biomesight-specific database is complete, an option for generating (offline-only) personalized suggestions will be added to the Microbiome Prescription website.

Probiotics Suggestions

The following are based on a simplified algorithm using R2 data for Biomesight. These are tentative numbers subject to future refinements. Bacteria listed are only for probiotics detected with Biomesight tests. Probiotics include some that are available only in some countries and some that are pending approval for retail sale.

• Good Count: Number of bacteria expected to shift in desired direction
• Bad Count: Number of bacteria expected to shift in wrong direction
• Impact: Estimator of impact based on Chi-2, Slope and R2 vectors

Blautia hansenii	194.64	19	0
Blautia wexlerae	147.38	8	0
Faecalibacterium prausnitzii	74.57	10	3
Enterococcus faecalis	19.48	34	44
Lactobacillus johnsonii	9.37	29	62
Streptococcus thermophilus	4.1	8	0
Escherichia coli	3.35	8	2
Segatella copri	2.83	2	4
Enterococcus faecium	2.16	10	45
Bifidobacterium breve	1.91	2	3
Heyndrickxia coagulans	1.89	7	39
Bifidobacterium longum	1.7	2	2
Bifidobacterium adolescentis	1.46	2	0
Enterococcus durans	1.4	27	30
Bifidobacterium bifidum	0.48	3	2
Bifidobacterium catenulatum	0.27	3	4
Lacticaseibacillus paracasei	0.2	5	14
Bifidobacterium animalis	0.16	1	1
Leuconostoc mesenteroides	0.09	11	26
Limosilactobacillus fermentum	0.04	9	29
Veillonella atypica	-0.03	2	4
Ligilactobacillus salivarius	-0.11	5	21
Lacticaseibacillus casei	-0.13	1	12
Lacticaseibacillus rhamnosus	-0.23	1	14
Lactiplantibacillus plantarum	-0.26	0	12
Limosilactobacillus reuteri	-0.47	24	31
Akkermansia muciniphila	-0.51	3	15
Lactobacillus acidophilus	-0.7	5	42
Lactobacillus crispatus	-0.95	1	43
Odoribacter laneus	-0.96	0	3
Bacillus subtilis	-0.99	29	55
Limosilactobacillus vaginalis	-1.33	21	70
Parabacteroides distasonis	-1.35	0	1
Bacteroides uniformis	-2.4	3	3
Lactobacillus jensenii	-2.56	6	82
Bifidobacterium pseudocatenulatum	-3.11	5	43
Lactobacillus helveticus	-5.05	35	100
Bacteroides thetaiotaomicron	-5.17	2	3
Clostridium butyricum	-5.21	0	35
Pediococcus acidilactici	-9.29	40	51
Parabacteroides goldsteinii	-9.3	0	12

Biomesight microbiome samples. The methodology for data acquisition is outlined in New Standards for Microbiome Analysis?. Self-described: Official Diagnosis: Mast Cell Dysfunction​

Tables have been refined to display only genus- and species-level taxa, the 20 most prominent entries per group, and associations achieving statistical significance (P < 0.01).​

The following sections provide the processed data, accompanied by guidance on interpretation and application. Counts of significant bacterial taxa are included, reflecting the application of non-standard but rigorously validated statistical approaches to extensive sample and reference populations, where statistical power derives from dataset scale.

Significance	Genus
p < 0.01	131
p < 0.001	118
p < 0.0001	106
p < 0.00001	94

Averages and Medians

I prefer medians over averages. Medians are the values where half of the people have less and half has more. If the data was a bell-curve, then the values will almost be the same… with bacteria that happens rarely. Look at the bacterua below, we see that for some the average is above and the median below. Should one increase or decrease this bacteria?

If symptom median is higher than reference median, it means there is more of this bacteria. If lower, then less. This ignores how often the bacteria is seen (we average only over reports).
IMHO using average value instead of median will often result in a worse situation for the patient

tax_name	Rank	Symptom Avarage	Reference Average	Symptom Median	Reference Median
Phocaeicola dorei	species	4.464	2.872	0.399	0.92
Roseburia	genus	3.065	2.833	1.786	2.058
Sutterella	genus	1.711	1.643	1.259	1.022
Parabacteroides merdae	species	0.558	0.75	0.306	0.09
Clostridium	genus	1.977	1.856	1.363	1.566
Bacteroides thetaiotaomicron	species	1.754	1.057	0.464	0.659
Coprococcus	genus	1.271	1.435	0.73	0.597
Mediterraneibacter	genus	1.174	0.706	0.279	0.386
Bacteroides caccae	species	1.451	0.864	0.29	0.19
Bacteroides cellulosilyticus	species	1.26	0.845	0.076	0.158
Lachnospira pectinoschiza	species	0.67	0.663	0.334	0.257
Blautia obeum	species	0.593	0.572	0.233	0.303
Bilophila	genus	0.363	0.35	0.211	0.272
Hathewaya histolytica	species	0.426	0.275	0.156	0.205
Hathewaya	genus	0.427	0.276	0.156	0.205
Sutterella wadsworthensis	species	0.845	0.655	0.058	0.011
Veillonella criceti	species	0.279	0.237	0.124	0.168
Bacteroides rodentium	species	0.338	0.393	0.187	0.231
Akkermansia	genus	1.582	1.353	0.053	0.011
Akkermansia muciniphila	species	1.582	1.354	0.053	0.011

Bacteria Incidence – How often is it reported

The common sense belief is that if a bacteria is reported more often, then the amount should be higher. This is often not true. The microbiome is a complex thing.

Nothing found that was significant

More or Less often based on Symptom Median All Incidence

This is a little more complex to understand. If we compute the mid point for people with the symptom, then if the bacteria was not involved then half of the reference should be above this value and half below this value. If not, it means that the symptom tends to over or under growth.

tax_name	Rank	Symptom Median	Odds Ratio	Chi2	Below	Above
Sulfobacillus acidophilus	species	0.002	0.39	10.8	84	33
Sulfobacillus	genus	0.002	0.39	10.8	84	33
Caldanaerobacter hydrothermalis	species	0.002	0.43	9.8	96	41
Caldanaerobacter	genus	0.002	0.43	9.8	96	41
Desulfotomaculum defluvii	species	0.003	0.56	8.1	1032	578
Pelagicoccus	genus	0.002	0.57	7.4	859	490
Alkalibacterium	genus	0.003	0.57	7.4	907	518
Hydrogenophilus	genus	0.003	0.58	7.3	1162	670
Sporotomaculum syntrophicum	species	0.003	0.59	6.8	1138	668

More or Less often based on Reference Median All Incidence

This is like the above, but with a different line in the sand. Instead of the median of those with the condition, we use the median of the reference set.

tax_name	Rank	Reference Median	Odds Ratio	Chi2	Below	Above
Nostoc	genus	0.003	0.34	295.7	1214	408
Bacillus	genus	0.006	0.43	277.1	1954	837
Erysipelothrix	genus	0.018	0.47	250.1	2346	1111
Psychrobacter	genus	0.003	0.39	239.8	1254	492
Sharpea	genus	0.025	0.4	233	1278	514
Methylobacillus glycogenes	species	0.003	0.4	232.7	1286	519
Sharpea azabuensis	species	0.025	0.41	226.8	1264	514
Methylobacillus	genus	0.003	0.42	219.3	1287	537
Erysipelothrix muris	species	0.017	0.5	218.7	2274	1130
Candidatus Tammella caduceiae	species	0.003	0.41	205.7	1155	478
Paenibacillus	genus	0.003	0.39	205.3	1016	398
Candidatus Tammella	genus	0.003	0.42	200.1	1170	494
[Ruminococcus] torques	species	0.04	0.51	189.3	1921	971
Holdemania	genus	0.028	0.53	187	2315	1225
Streptococcus oralis	species	0.003	0.47	185.8	1453	686
Amedibacillus dolichus	species	0.024	0.5	184.4	1759	879
Amedibacillus	genus	0.024	0.5	184	1758	879
Haemophilus parainfluenzae	species	0.01	1.91	173.7	1023	1952
Haemophilus	genus	0.01	1.89	170.3	1035	1959
Luteolibacter	genus	0.015	0.46	169.7	1177	541

More or Less often based on Symptom Median High Incidence

Above we see that many of the top bacteria identified are sparse, that is not reported often. We then restrict them to those that occur above 50% or the time.

Nothing found that was significant

More or Less often based on Reference Median High Incidence

Above we see that many of the top bacteria identified are sparse, that is not reported often. We then restrict them to those that occur above 50% or the time.

tax_name	Rank	Reference Median Freq	Odds Ratio	Chi2	Below	Above
Nostoc	genus	0.003	0.34	295.7	1214	408
Bacillus	genus	0.006	0.43	277.1	1954	837
Erysipelothrix	genus	0.018	0.47	250.1	2346	1111
Psychrobacter	genus	0.003	0.39	239.8	1254	492
Sharpea	genus	0.025	0.4	233	1278	514
Methylobacillus glycogenes	species	0.003	0.4	232.7	1286	519
Sharpea azabuensis	species	0.025	0.41	226.8	1264	514
Methylobacillus	genus	0.003	0.42	219.3	1287	537
Erysipelothrix muris	species	0.017	0.5	218.7	2274	1130
Candidatus Tammella caduceiae	species	0.003	0.41	205.7	1155	478
Paenibacillus	genus	0.003	0.39	205.3	1016	398
Candidatus Tammella	genus	0.003	0.42	200.1	1170	494
[Ruminococcus] torques	species	0.04	0.51	189.3	1921	971
Holdemania	genus	0.028	0.53	187	2315	1225
Streptococcus oralis	species	0.003	0.47	185.8	1453	686
Amedibacillus dolichus	species	0.024	0.5	184.4	1759	879
Amedibacillus	genus	0.024	0.5	184	1758	879
Haemophilus parainfluenzae	species	0.01	1.91	173.7	1023	1952
Haemophilus	genus	0.01	1.89	170.3	1035	1959
Luteolibacter	genus	0.015	0.46	169.7	1177	541

Summary

A large number of bacterial taxa exhibit shifts with P < 0.01 in association with this condition. The subsequent challenge is determining how to modulate these taxa, since the volume of candidates exceeds what most individuals can practically consider. Moreover, for many of the taxa identified, there is no published evidence in the U.S. National Library of Medicine describing how to alter their abundance.

A deep optimization model, such as the one implemented on the Microbiome Taxa R2 site, can be used to inform probiotic selection. This model provides coverage for each identified taxon and infers which probiotics are most likely to shift their levels. Its output may then be integrated with more conventional recommendations derived from literature indexed in the U.S. National Library of Medicine where such evidence exists, with the two recommendation sets reconciled by giving priority to probiotic-based suggestions.

Development of a dedicated database based on Biomesight samples is in progress. The current model uses data contributed by PrecisionBiome, and datasets generated with differing laboratory processing pipelines cannot be safely combined, as discussed in The taxonomy nightmare before Christmas…. Once the Biomesight-specific database is complete, an option for generating (offline-only) personalized suggestions will be added to the Microbiome Prescription website.

Probiotics Suggestions

The following are based on a simplified algorithm using R2 data for Biomesight. These are tentative numbers subject to future refinements. Bacteria listed are only for probiotics detected with Biomesight tests. Probiotics include some that are available only in some countries and some that are pending approval for retail sale.

• Good Count: Number of bacteria expected to shift in desired direction
• Bad Count: Number of bacteria expected to shift in wrong direction
• Impact: Estimator of impact based on Chi-2, Slope and R2 vectors

Probiotic Species	Impact	Good Count	Bad Count
Akkermansia muciniphila	36.08	18	6
Segatella copri	32.72	5	2
Bifidobacterium breve	21.31	8	5
Bifidobacterium longum	19.06	8	6
Bifidobacterium adolescentis	14.38	8	7
Lactobacillus helveticus	9.11	48	27
Streptococcus thermophilus	7.97	8	2
Lactobacillus johnsonii	7.28	22	26
Bifidobacterium bifidum	4.35	7	2
Bifidobacterium catenulatum	4.24	8	0
Parabacteroides goldsteinii	4.11	5	9
Bifidobacterium animalis	1.97	7	0
Lactococcus lactis	1.1	6	2
Veillonella atypica	1.02	11	3
Clostridium butyricum	0.97	7	9
Limosilactobacillus vaginalis	0.86	20	29
Odoribacter laneus	0.76	2	0
Enterococcus durans	0.7	11	20
Limosilactobacillus fermentum	0.11	2	15
Leuconostoc mesenteroides	-0.07	3	7
Lacticaseibacillus paracasei	-0.07	2	9
Lacticaseibacillus rhamnosus	-0.09	0	2
Bifidobacterium pseudocatenulatum	-0.1	5	6
Heyndrickxia coagulans	-0.1	3	9
Ligilactobacillus salivarius	-0.14	1	6
Lactobacillus crispatus	-0.15	3	7
Lactiplantibacillus plantarum	-0.19	0	4
Lactiplantibacillus pentosus	-0.21	0	4
Lacticaseibacillus casei	-0.21	0	7
Lactobacillus acidophilus	-0.22	8	12
Bacillus subtilis	-0.26	8	27
Limosilactobacillus reuteri	-0.42	5	16
Lactobacillus jensenii	-1.41	15	29
Pediococcus acidilactici	-1.86	17	32
Enterococcus faecium	-2.58	7	22
Enterococcus faecalis	-9.41	28	50
Parabacteroides distasonis	-9.5	9	4
Blautia wexlerae	-13.3	1	3
Escherichia coli	-24.86	1	12
Blautia hansenii	-26.65	7	6
Faecalibacterium prausnitzii	-116.83	3	3
Bacteroides uniformis	-167.73	1	11
Bacteroides thetaiotaomicron	-178.11	1	11

Comments on this Condition

Two of the above sections reported nothing significant found. This implies that the microbiome plays a secondary role. The bacteria shifts are more likely consequences of the condition than triggers of the condition. Regardless, there is a potential that the above probiotics may modify the severity of the condition.

It is unclear if the shifts are due to anti-histamine and other drugs usage.

This document presents the results of statistical analysis on symptoms from viable, self-annotated Biomesight microbiome samples. The methodology for data acquisition is outlined in New Standards for Microbiome Analysis?.​

Tables have been refined to display only genus- and species-level taxa, the 20 most prominent entries per group, and associations achieving statistical significance (P < 0.01).​

The following sections provide the processed data, accompanied by guidance on interpretation and application. Counts of significant bacterial taxa are included, reflecting the application of non-standard but rigorously validated statistical approaches to extensive sample and reference populations, where statistical power derives from dataset scale.

Significance	Genus
p < 0.01	196
p < 0.001	172
p < 0.0001	154
p < 0.00001	140

Averages and Medians

I prefer medians over averages. Medians are the values where half of the people have less and half has more. If the data was a bell-curve, then the values will almost be the same… with bacteria that happens rarely. Look at Bifidobacterium below, we see that the average is above and the median below.

If symptom median is higher than reference median, it means there is more of this bacteria. If lower, then less. This ignores how often the bacteria is seen (we average only over reports).

tax_name	Rank	Symptom Avarage	Reference Average	Symptom Median	Reference Median
Bacteroides	genus	28.802	25.926	24.224	27.059
Phocaeicola	genus	12.357	10.786	9.314	11.306
Phocaeicola vulgatus	species	7.043	5.751	3.394	4.929
Bacteroides uniformis	species	2.909	2.723	1.553	1.958
Bacteroides thetaiotaomicron	species	1.234	1.065	0.458	0.734
Coprococcus	genus	1.23	1.44	0.737	0.552
Roseburia faecis	species	0.969	1.217	0.577	0.455
Bilophila	genus	0.417	0.347	0.207	0.319
Bifidobacterium	genus	0.534	0.953	0.131	0.035
Bacteroides stercoris	species	2.066	1.543	0.033	0.123
Blautia coccoides	species	0.776	0.917	0.592	0.504
Bilophila wadsworthia	species	0.395	0.339	0.197	0.281
Mediterraneibacter	genus	0.885	0.708	0.278	0.326
Butyricimonas	genus	0.217	0.186	0.108	0.154
Hathewaya	genus	0.314	0.277	0.155	0.201
Hathewaya histolytica	species	0.314	0.277	0.155	0.201
Bacteroides rodentium	species	0.435	0.39	0.186	0.231
Bifidobacterium longum	species	0.237	0.326	0.051	0.012
Lachnobacterium	genus	0.197	0.327	0.075	0.041
Bacteroides stercorirosoris	species	0.234	0.191	0.135	0.164

Bacteria Incidence – How often is it reported

The common sense belief is that if a bacteria is reported more often, then the amount should be higher. This is often not true. The microbiome is a complex thing. Look at Bacteroides uniformis below, we see that the average is above and the median below

tax_name	Rank	Incidence Odds Ratio	Chi2	Symptoms %	Reference %
Bifidobacterium breve	species	0.64	7.8	26.6	41.4
Anaerococcus hydrogenalis	species	1.66	7.2	18.2	11

More or Less often based on Symptom Median All Incidence

This is a little more complex to understand. If we compute the mid point for people with the symptom, then if the bacteria was not involved then half of the reference should be above this value and half below this value. If not, it means that the symptom tends to over or under growth.

tax_name	Rank	Symptom Median	Odds Ratio	Chi2	Below	Above
Niabella aurantiaca	species	0.002	0.34	44.3	544	183
Psychroflexus	genus	0.002	0.31	43.7	357	111
Psychroflexus gondwanensis	species	0.002	0.31	43.7	357	111
Rickettsia marmionii Stenos et al. 2005	species	0.002	0.33	41.6	398	131
Psychrobacter glacialis	species	0.002	0.37	38.6	660	246
Niabella	genus	0.002	0.38	36	584	222
Chromatium	genus	0.002	0.38	34.4	517	198
Chromatium weissei	species	0.002	0.38	34.2	516	198
Lentibacillus	genus	0.002	0.38	34	510	196
Lentibacillus salinarum	species	0.002	0.38	33.7	494	190
Viridibacillus neidei	species	0.002	0.38	33.3	469	180
Thermoanaerobacterium	genus	0.002	0.41	30	483	196
Thiomicrospira	genus	0.002	0.39	29.3	335	130
Sporosarcina pasteurii	species	0.002	0.41	29.2	439	178
Thiorhodococcus	genus	0.002	0.42	29.1	578	243
Thermoanaerobacterium islandicum	species	0.002	0.41	29	476	196
Syntrophomonas sapovorans	species	0.002	0.42	29	534	223
Sporosarcina	genus	0.002	0.42	27.5	443	185
Thermodesulfovibrio thiophilus	species	0.002	0.44	25.5	536	237
Oenococcus	genus	0.002	0.45	24.8	604	273

More or Less often based on Reference Median All Incidence

This is like the above, but with a different line in the sand. Instead of the median of those with the condition, we use the median of the reference set.

tax_name	Rank	Reference Median	Odds Ratio	Chi2	Below	Above
Bifidobacterium longum	species	0.012	2.27	260.2	870	1971
Bifidobacterium	genus	0.035	2.04	241.7	1266	2586
Methylobacillus glycogenes	species	0.003	0.4	232.5	1250	497
Methylobacillus	genus	0.003	0.41	217.2	1249	516
Corynebacterium	genus	0.0085	0.42	201.9	1163	486
Bilophila	genus	0.3185	0.55	162.8	2247	1236
Erysipelothrix muris	species	0.015	0.55	161	2153	1179
Psychrobacter glacialis	species	0.002	0.37	156.5	660	246
Niabella aurantiaca	species	0.002	0.34	153.5	544	183
Methylonatrum	genus	0.004	0.53	145.2	1620	866
Methylonatrum kenyense	species	0.004	0.53	145.2	1620	866
Catonella morbi	species	0.01	0.56	144	1968	1099
Catonella	genus	0.01	0.56	141.4	1966	1104
Erysipelothrix	genus	0.0155	0.57	139.9	2151	1232
Niabella	genus	0.002	0.38	137	584	222
Megasphaera elsdenii	species	0.0045	0.41	132.7	640	260
Bacteroides thetaiotaomicron	species	0.734	0.6	130.1	2389	1422
Veillonella parvula	species	0.003	1.9	128.6	666	1266
Alkalithermobacter paradoxus	species	0.004	0.55	125.8	1537	853
Odoribacter denticanis	species	0.005	0.57	124.1	1642	928

More or Less often based on Symptom Median High Incidence

Above we see that many of the top bacteria identified are sparse, that is not reported often. We then restrict them to those that occur above 50% or the time.

tax_name	Rank	Symptom Median Freq	Odds Ratio	Chi2	Below	Above
Clostridium taeniosporum	species	0.003	0.61	11.2	1344	814
Dethiosulfovibrio	genus	0.004	0.66	7.6	1500	994
Tetragenococcus doogicus	species	0.003	0.67	7.2	1360	910

More or Less often based on Reference Median High Incidence

Above we see that many of the top bacteria identified are sparse, that is not reported often. We then restrict them to those that occur above 50% or the time.

tax_name	Rank	Reference Median Freq	Odds Ratio	Chi2	Below	Above
Bifidobacterium longum	species	0.012	2.27	260.2	870	1971
Bifidobacterium	genus	0.035	2.04	241.7	1266	2586
Methylobacillus glycogenes	species	0.003	0.4	232.5	1250	497
Methylobacillus	genus	0.003	0.41	217.2	1249	516
Corynebacterium	genus	0.0085	0.42	201.9	1163	486
Bilophila	genus	0.3185	0.55	162.8	2247	1236
Erysipelothrix muris	species	0.015	0.55	161	2153	1179
Psychrobacter glacialis	species	0.002	0.37	156.5	660	246
Niabella aurantiaca	species	0.002	0.34	153.5	544	183
Methylonatrum	genus	0.004	0.53	145.2	1620	866
Methylonatrum kenyense	species	0.004	0.53	145.2	1620	866
Catonella morbi	species	0.01	0.56	144	1968	1099
Catonella	genus	0.01	0.56	141.4	1966	1104
Erysipelothrix	genus	0.0155	0.57	139.9	2151	1232
Niabella	genus	0.002	0.38	137	584	222
Megasphaera elsdenii	species	0.0045	0.41	132.7	640	260
Bacteroides thetaiotaomicron	species	0.734	0.6	130.1	2389	1422
Veillonella parvula	species	0.003	1.9	128.6	666	1266
Alkalithermobacter paradoxus	species	0.004	0.55	125.8	1537	853
Odoribacter denticanis	species	0.005	0.57	124.1	1642	928

Summary

A large number of bacterial taxa exhibit shifts with P < 0.01 in association with this condition. The subsequent challenge is determining how to modulate these taxa, since the volume of candidates exceeds what most individuals can practically consider. Moreover, for many of the taxa identified, there is no published evidence in the U.S. National Library of Medicine describing how to alter their abundance.

A deep optimization model, such as the one implemented on the Microbiome Taxa R2 site, can be used to inform probiotic selection. This model provides coverage for each identified taxon and infers which probiotics are most likely to shift their levels. Its output may then be integrated with more conventional recommendations derived from literature indexed in the U.S. National Library of Medicine where such evidence exists, with the two recommendation sets reconciled by giving priority to probiotic-based suggestions.

Development of a dedicated database based on Biomesight samples is in progress. The current model uses data contributed by PrecisionBiome, and datasets generated with differing laboratory processing pipelines cannot be safely combined, as discussed in The taxonomy nightmare before Christmas…. Once the Biomesight-specific database is complete, an option for generating (offline-only) personalized suggestions will be added to the Microbiome Prescription website.

Probiotics Suggestions

The following are based on a simplified algorithm using R2 data for Biomesight. These are tentative numbers subject to future refinements. Bacteria listed are only for probiotics detected with Biomesight tests. Probiotics include some that are available only in some countries and some that are pending approval for retail sale.

• Good Count: Number of bacteria expected to shift in desired direction
• Bad Count: Number of bacteria expected to shift in wrong direction
• Impact: Estimator of impact based on Chi-2, Slope and R2 vectors

Some literature suggesting that the model’s suggestions are reasonable:

• Bifidobacterium breve Bif11 supplementation improves depression-related neurobehavioural and neuroinflammatory changes in the mouse. Neuropharmacology (Neuropharmacology ) Vol: 229 Issue: Pages: 109480 Pub: 2023 May 15 ePub: 2023 Mar 1 Authors Sushma G,Vaidya B,Sharma S,Devabattula G,Bishnoi M,Kondepudi KK,Sharma SS
• Heat-sterilized Bifidobacterium breve prevents depression-like behavior and interleukin-1ß expression in mice exposed to chronic social defeat stress. Brain, behavior, and immunity (Brain Behav Immun ) Vol: Issue: Pages: Pub: 2021 May 29 ePub: 2021 May 29 Authors Kosuge A,Kunisawa K,Arai S,Sugawara Y,Shinohara K,Iida T,Wulaer B,Kawai T,Fujigaki H,Yamamoto Y,Saito K,Nabeshima T,Mouri A
• Bifidobacterium breve BB05 alleviates depressive symptoms in mice via the AKT/mTOR pathway.
  Frontiers in nutrition (Front Nutr ) Vol: 12 Issue: Pages: 1529566 Pub: 2025 ePub: 2025 Jan 30 Authors Pan Y,Huang Q,Liang Y,Xie Y,Tan F,Long X
• Lipid and Energy Metabolism of the Gut Microbiota Is Associated with the Response to Probiotic Bifidobacterium breve Strain for Anxiety and Depressive Symptoms in Schizophrenia.
  Journal of personalized medicine (J Pers Med ) Vol: 11 Issue: 10 Pages: Pub: 2021 Sep 30 ePub: 2021 Sep 30 Authors Yamamura R,Okubo R,Katsumata N,Odamaki T,Hashimoto N,Kusumi I,Xiao J,Matsuoka YJ
• Towards a psychobiotic therapy for depression: Bifidobacterium breve CCFM1025 reverses chronic stress-induced depressive symptoms and gut microbial abnormalities in mice. Neurobiology of stress (Neurobiol Stress ) Vol: 12 Issue: Pages: 100216 Pub: 2020 May ePub: 2020 Mar 20 Authors Tian P,O’Riordan KJ,Lee YK,Wang G,Zhao J,Zhang H,Cryan JF,Chen W
  

Probiotic Species	Impact	Good Count	Bad Count
Faecalibacterium prausnitzii	256.62	8	0
Blautia hansenii	181.61	22	1
Bifidobacterium breve	89.17	15	0
Bifidobacterium longum	79.15	15	2
Blautia wexlerae	59.88	7	0
Bifidobacterium adolescentis	58.57	14	2
Segatella copri	41.79	3	0
Bifidobacterium bifidum	17.98	11	2
Bifidobacterium catenulatum	15.57	8	0
Escherichia coli	14.66	11	0
Enterococcus faecalis	14.27	42	46
Akkermansia muciniphila	11.57	12	9
Lactobacillus helveticus	9.86	43	53
Bifidobacterium animalis	8.17	7	0
Enterococcus faecium	5.05	14	34
Streptococcus thermophilus	2.3	4	0
Enterococcus durans	2.14	25	23
Veillonella atypica	1.34	12	0
Clostridium butyricum	0.67	13	26
Bacillus subtilis	0.51	28	33
Lacticaseibacillus paracasei	0.23	13	5
Lactococcus lactis	0.15	2	5
Limosilactobacillus fermentum	0.12	10	17
Lactiplantibacillus pentosus	0.1	6	4
Heyndrickxia coagulans	0.04	8	14
Ligilactobacillus salivarius	-0.05	2	10
Lactiplantibacillus plantarum	-0.08	3	6
Lacticaseibacillus casei	-0.17	1	10
Lacticaseibacillus rhamnosus	-0.18	2	5
Limosilactobacillus vaginalis	-0.22	25	51
Leuconostoc mesenteroides	-0.26	4	13
Bifidobacterium pseudocatenulatum	-0.35	7	24
Lactobacillus crispatus	-0.63	2	31
Limosilactobacillus reuteri	-0.78	17	22
Lactobacillus acidophilus	-0.83	5	33
Odoribacter laneus	-1.78	0	3
Lactobacillus jensenii	-3.23	18	55
Pediococcus acidilactici	-7.22	26	33
Lactobacillus johnsonii	-13.52	36	39
Parabacteroides goldsteinii	-21.27	0	13
Parabacteroides distasonis	-27.34	0	11
Bacteroides uniformis	-285.55	0	11
Bacteroides thetaiotaomicron	-294.72	0	10

This document presents the results of statistical analysis on symptoms from viable, self-annotated Biomesight microbiome samples. The methodology for data acquisition is outlined in New Standards for Microbiome Analysis?.​

Tables have been refined to display only genus- and species-level taxa, the 20 most prominent entries per group, and associations achieving statistical significance (P < 0.01).​

The following sections provide the processed data, accompanied by guidance on interpretation and application. Counts of significant bacterial taxa are included, reflecting the application of non-standard but rigorously validated statistical approaches to extensive sample and reference populations, where statistical power derives from dataset scale.

Significance	Genus
p < 0.01	149
p < 0.001	121
p < 0.0001	96
p < 0.00001	79

Averages and Medians

I prefer medians over averages. Medians are the values where half of the people have less and half has more. If the data was a bell-curve, then the values will almost be the same… with bacteria that happens rarely. Look at Bacteroides uniformis below, we see that the average is above and the median below.

If symptom median is higher than reference median, it means there is more of this bacteria. If lower, then less. This ignores how often the bacteria is seen (we average only over reports).

tax_name	Rank	Symptom Avarage	Reference Average	Symptom Median	Reference Median
Bacteroides	genus	28.504	25.674	23.872	27.629
Bacteroides uniformis	species	3.078	2.678	1.51	1.977
Coprococcus	genus	1.258	1.458	0.75	0.539
Bifidobacterium	genus	0.677	0.974	0.135	0.064
Bacteroides cellulosilyticus	species	1.06	0.822	0.073	0.126
Bilophila	genus	0.397	0.343	0.206	0.252
Bilophila wadsworthia	species	0.385	0.335	0.196	0.239
Alkaliphilus	genus	0.251	0.299	0.07	0.041
Alkaliphilus crotonatoxidans	species	0.245	0.291	0.065	0.037
Collinsella	genus	0.146	0.188	0.058	0.036
Bifidobacterium longum	species	0.221	0.338	0.052	0.03
Bacteroides stercorirosoris	species	0.216	0.19	0.134	0.156
Collinsella aerofaciens	species	0.138	0.175	0.056	0.036
Caloramator mitchellensis	species	0.798	0.873	0.054	0.035
Anaerotruncus	genus	0.195	0.186	0.136	0.155
Bacteroides salyersiae	species	0.262	0.367	0.022	0.004
Bacteroides faecis	species	0.145	0.118	0.055	0.071
Anaerotruncus colihominis	species	0.184	0.174	0.133	0.147
Oxalobacter	genus	0.036	0.03	0.018	0.027
Luteibacter anthropi	species	0.051	0.08	0.016	0.009

Bacteria Incidence – How often is it reported

The common sense belief is that if a bacteria is reported more often, then the amount should be higher. This is often not true. The microbiome is a complex thing. Look at Bacteroides uniformis below, we see that the average is above and the median below

tax_name	Rank	Incidence Odds Ratio	Chi2	Symptoms %	Reference %
Bifidobacterium scardovii	species	0.71	6.6	12.3	17.2

More or Less often based on Symptom Median All Incidence

This is a little more complex to understand. If we compute the mid point for people with the symptom, then if the bacteria was not involved then half of the reference should be above this value and half below this value. If not, it means that the symptom tends to over or under growth.

tax_name	Rank	Symptom Median	Odds Ratio	Chi2	Below	Above
Niabella aurantiaca	species	0.002	0.34	79.6	484	165
Psychrobacter glacialis	species	0.002	0.37	78.5	594	218
Niabella	genus	0.002	0.39	66	516	201
Viridibacillus neidei	species	0.002	0.38	61.6	426	163
Actinopolyspora	genus	0.002	0.4	61.5	489	195
Chromatium	genus	0.002	0.4	61.4	473	187
Chromatium weissei	species	0.002	0.4	61.1	472	187
Lentibacillus	genus	0.002	0.4	58.9	459	184
Lentibacillus salinarum	species	0.002	0.41	56.7	444	180
Thermoanaerobacterium	genus	0.002	0.42	53.1	438	183
Thiorhodococcus	genus	0.002	0.44	53.1	524	229
Thermoanaerobacterium islandicum	species	0.002	0.42	51.3	432	183
Syntrophomonas sapovorans	species	0.002	0.44	50.8	482	211
Thermodesulfovibrio thiophilus	species	0.002	0.44	49.5	481	213
Thermodesulfovibrio	genus	0.002	0.47	46.8	552	258
Desulfofundulus	genus	0.002	0.45	46.7	449	201
Helicobacter suncus	species	0.002	0.49	46.3	669	325
Oenococcus	genus	0.002	0.47	45.1	529	249
Vagococcus penaei	species	0.003	0.49	36.1	427	211
Viridibacillus	genus	0.002	0.5	34.8	437	220

More or Less often based on Reference Median All Incidence

This is like the above, but with a different line in the sand. Instead of the median of those with the condition, we use the median of the reference set.

tax_name	Rank	Reference Median	Odds Ratio	Chi2	Below	Above
Psychrobacter glacialis	species	0.002	0.37	144.3	594	218
Niabella aurantiaca	species	0.002	0.34	134.5	484	165
Niabella	genus	0.002	0.39	116.9	516	201
Bacteroides heparinolyticus	species	0.003	0.49	111.1	858	423
Actinopolyspora	genus	0.002	0.4	107.5	489	195
Chromatium	genus	0.002	0.4	106	473	187
Chromatium weissei	species	0.002	0.4	105.4	472	187
Viridibacillus neidei	species	0.002	0.38	102	426	163
Lentibacillus	genus	0.002	0.4	100.9	459	184
Thiorhodococcus	genus	0.002	0.44	96.8	524	229
Lentibacillus salinarum	species	0.002	0.41	96.2	444	180
Helicobacter suncus	species	0.002	0.49	94.8	669	325
Thermoanaerobacterium	genus	0.002	0.42	90.2	438	183
Syntrophomonas sapovorans	species	0.002	0.44	89.9	482	211
Thermodesulfovibrio	genus	0.002	0.47	88.2	552	258
Thermodesulfovibrio thiophilus	species	0.002	0.44	87.8	481	213
Thermoanaerobacterium islandicum	species	0.002	0.42	87	432	183
Oenococcus	genus	0.002	0.47	83.9	529	249
Hydrogenophilus	genus	0.003	0.58	81.2	1019	589
Desulfofundulus	genus	0.002	0.45	81	449	201

More or Less often based on Symptom Median High Incidence

Above we see that many of the top bacteria identified are sparse, that is not reported often. We then restrict them to those that occur above 50% or the time.

tax_name	Rank	Symptom Median Freq	Odds Ratio	Chi2	Below	Above
Clostridium taeniosporum	species	0.003	0.62	24.8	1203	751
Dethiosulfovibrio	genus	0.004	0.68	17.5	1353	917
Tetragenococcus doogicus	species	0.003	0.69	15.8	1214	836
Hydrocarboniphaga daqingensis	species	0.004	0.71	13.8	1453	1032
Mycoplasmopsis	genus	0.005	0.72	13.3	1600	1147
Pediococcus	genus	0.004	0.76	8.7	1168	886
Propionispora	genus	0.005	0.77	7.9	1374	1060
Propionispora hippei	species	0.005	0.77	7.9	1374	1060
Tetragenococcus	genus	0.003	0.78	6.7	1186	931

More or Less often based on Reference Median High Incidence

Above we see that many of the top bacteria identified are sparse, that is not reported often. We then restrict them to those that occur above 50% or the time.

tax_name	Rank	Reference Median Freq	Odds Ratio	Chi2	Below	Above
Psychrobacter glacialis	species	0.002	0.37	144.3	594	218
Niabella aurantiaca	species	0.002	0.34	134.5	484	165
Niabella	genus	0.002	0.39	116.9	516	201
Bacteroides heparinolyticus	species	0.003	0.49	111.1	858	423
Actinopolyspora	genus	0.002	0.4	107.5	489	195
Chromatium	genus	0.002	0.4	106	473	187
Chromatium weissei	species	0.002	0.4	105.4	472	187
Viridibacillus neidei	species	0.002	0.38	102	426	163
Lentibacillus	genus	0.002	0.4	100.9	459	184
Thiorhodococcus	genus	0.002	0.44	96.8	524	229
Lentibacillus salinarum	species	0.002	0.41	96.2	444	180
Helicobacter suncus	species	0.002	0.49	94.8	669	325
Thermoanaerobacterium	genus	0.002	0.42	90.2	438	183
Syntrophomonas sapovorans	species	0.002	0.44	89.9	482	211
Thermodesulfovibrio	genus	0.002	0.47	88.2	552	258
Thermodesulfovibrio thiophilus	species	0.002	0.44	87.8	481	213
Thermoanaerobacterium islandicum	species	0.002	0.42	87	432	183
Oenococcus	genus	0.002	0.47	83.9	529	249
Hydrogenophilus	genus	0.003	0.58	81.2	1019	589
Desulfofundulus	genus	0.002	0.45	81	449	201

Summary

A large number of bacterial taxa exhibit shifts with P < 0.01 in association with this condition. The subsequent challenge is determining how to modulate these taxa, since the volume of candidates exceeds what most individuals can practically consider. Moreover, for many of the taxa identified, there is no published evidence in the U.S. National Library of Medicine describing how to alter their abundance.

A deep optimization model, such as the one implemented on the Microbiome Taxa R2 site, can be used to inform probiotic selection. This model provides coverage for each identified taxon and infers which probiotics are most likely to shift their levels. Its output may then be integrated with more conventional recommendations derived from literature indexed in the U.S. National Library of Medicine where such evidence exists, with the two recommendation sets reconciled by giving priority to probiotic-based suggestions.

Development of a dedicated database based on Biomesight samples is in progress. The current model uses data contributed by PrecisionBiome, and datasets generated with differing laboratory processing pipelines cannot be safely combined, as discussed in The taxonomy nightmare before Christmas…. Once the Biomesight-specific database is complete, an option for generating (offline-only) personalized suggestions will be added to the Microbiome Prescription website.

Probiotics Suggestions

The following are based on a simplified algorithm using R2 data for Biomesight. These are tentative numbers subject to future refinements. Bacteria listed are only for probiotics detected with Biomesight tests. Probiotics include some that are available only in some countries and some that are pending approval for retail sale.

• Good Count: Number of bacteria expected to shift in desired direction
• Bad Count: Number of bacteria expected to shift in wrong direction
• Impact: Estimator of impact based on Chi-2, Slope and R2 vectors

Probiotic Species	Impact	Good Count	Bad Count
Bifidobacterium breve	53.54	16	0
Blautia hansenii	52.64	12	0
Bifidobacterium longum	48.23	19	0
Segatella copri	45.23	5	0
Bifidobacterium adolescentis	35.47	15	0
Lactobacillus helveticus	20.58	75	11
Akkermansia muciniphila	18.74	16	1
Faecalibacterium prausnitzii	15.59	3	0
Bifidobacterium bifidum	10.9	16	0
Bifidobacterium catenulatum	10.1	15	0
Enterococcus faecalis	9.15	55	10
Escherichia coli	6.29	8	0
Bifidobacterium animalis	4.92	7	0
Lactobacillus johnsonii	3.79	51	5
Enterococcus faecium	3.77	28	3
Enterococcus durans	3.18	39	1
Streptococcus thermophilus	1.64	3	0
Blautia wexlerae	1.56	1	0
Limosilactobacillus vaginalis	1.06	34	6
Lactobacillus jensenii	0.85	41	10
Limosilactobacillus reuteri	0.52	23	6
Clostridium butyricum	0.49	19	3
Bifidobacterium pseudocatenulatum	0.37	23	0
Bacillus amyloliquefaciens group	0.31	44	11
Lacticaseibacillus paracasei	0.31	10	1
Lactobacillus acidophilus	0.27	24	1
Veillonella atypica	0.27	6	0
Heyndrickxia coagulans	0.26	20	2
Lactobacillus crispatus	0.24	17	2
Ligilactobacillus salivarius	0.16	9	0
Lacticaseibacillus casei	0.14	7	2
Lactiplantibacillus pentosus	0.14	5	0
Bacillus subtilis group	0.13	22	4
Limosilactobacillus fermentum	0.12	13	5
Lactococcus lactis	0.12	5	0
Leuconostoc mesenteroides	0.11	14	1
Lactiplantibacillus plantarum	0.1	4	1
Lacticaseibacillus rhamnosus	0.09	4	0
Pediococcus	0.07	4	0
Bacillus	0.07	2	0
Bacillus subtilis	-0.32	31	12
Pediococcus acidilactici	-2.33	28	12
Parabacteroides goldsteinii	-13.74	0	8
Parabacteroides distasonis	-27.7	0	8
Bacteroides uniformis	-246.88	0	13
Bacteroides thetaiotaomicron	-265.51	0	14

​