research article systematic analysis of the association...

Research ArticleSystematic Analysis of the Association between GutFlora and Obesity through High-Throughput Sequencing andBioinformatics Approaches

Chih-Min Chiu1 Wei-Chih Huang1 Shun-Long Weng1234 Han-Chi Tseng5

Chao Liang6 Wei-Chi Wang6 Ting Yang1 Tzu-Ling Yang1 Chen-Tsung Weng6

Tzu-Hao Chang7 and Hsien-Da Huang18910

1 Institute of Bioinformatics and Systems Biology National Chiao Tung University Hsinchu 300 Taiwan2Department of Obstetrics and Gynecology Hsinchu Mackay Memorial Hospital Hsinchu 300 Taiwan3Mackay Medicine Nursing and Management College Taipei 104 Taiwan4Department of Medicine Mackay Medical College New Taipei City 251 Taiwan5 Tseng Han-Chi General Hospital Nantou 542 Taiwan6Health GeneTech Corporation Taoyuan 330 Taiwan7Graduate Institute of Biomedical Informatics Taipei Medical University Taipei 110 Taiwan8Department of Biological Science and Technology National Chiao Tung University Hsinchu 300 Taiwan9Center for Bioinformatics Research National Chiao Tung University Hsinchu 300 Taiwan10Department of Biomedical Science and Environmental Biology Kaohsiung Medical University Kaohsiung 807 Taiwan

Correspondence should be addressed to Tzu-Hao Chang kevinchangtmuedutw andHsien-DaHuang bryanmailnctuedutw

Received 11 April 2014 Accepted 27 June 2014 Published 14 August 2014

Academic Editor Li-Ching Wu

Copyright copy 2014 Chih-Min Chiu et alThis is an open access article distributed under the Creative CommonsAttribution Licensewhich permits unrestricted use distribution and reproduction in any medium provided the original work is properly cited

Eighty-one stool samples from Taiwanese were collected for analysis of the association between the gut flora and obesity Thesupervised analysis showed that the most abundant genera of bacteria in normal samples (from people with a body mass index(BMI) le 24) were Bacteroides (277) Prevotella (194) Escherichia (12) Phascolarctobacterium (39) and Eubacterium (35)Themost abundant genera of bacteria in case samples (with a BMIge 27)wereBacteroides (29)Prevotella (21)Escherichia (74)Megamonas (51) and Phascolarctobacterium (38) A principal coordinate analysis (PCoA) demonstrated that normal sampleswere clustered more compactly than case samples An unsupervised analysis demonstrated that bacterial communities in the gutwere clustered into twomain groups N-like and OB-like groups Remarkably most normal samples (78) were clustered in the N-like group andmost case samples (81) were clustered in theOB-like group (Fisherrsquos119875 value = 161119864minus07)The results showed thatbacterial communities in the gut were highly associated with obesity This is the first study in Taiwan to investigate the associationbetween human gut flora and obesity and the results provide new insights into the correlation of bacteria with the rising trend inobesity

1 Background

Enterobacteria or gut microbiota in the human gastroin-testinal (GI) tract play important roles in the bodyrsquos func-tions For example they can regulate immune responses andmetabolic functions of the host [1ndash4] The gut microbiota

is frequently used to study the association between humanhealth and an individualrsquos lifestyle In 2011 three majorkinds of enterotypes consisting of Bacteroides Prevotellaand Ruminococcus [5] provided new perspectives to classifyindividuals Ruminococcus was reported to be an ambiguousenterotype and the Bacteroides and Prevotella enterotypes

Hindawi Publishing CorporationBioMed Research InternationalVolume 2014 Article ID 906168 10 pageshttpdxdoiorg1011552014906168

2 BioMed Research International

were associated with dietary habits [6] Animal protein andsaturated fats were highly correlated with the Bacteroidesenterotype Low meat intake and plant-based nutritionwith high carbohydrates were correlated with the Prevotellaenterotype

TheHumanMicrobiomeProject (HMP) [7]was launchedby the US National Institutes of Health in 2008 andunderstanding the relationship between human health andmicrobiota that live in or on the human body was recognizedas an important concept To decipher this relationship high-throughput sequencing also called next-generation sequenc-ing (NGS) supporting a large number of sequences can beused to sequence 16S ribosomal (r)RNA to construct complexmicrobial community profiles 16S rRNA is considered thestandard for studying microbial communities and assigningtaxonomy to bacteria Compared to conventional polymerasechain reaction- (PCR-) based or culture-based methods 16SrRNA sequencing by NGS can detect hundreds to thousandsof bacteria at one time and offer relative quantification of thebacteria Interactions between different bacterial communi-ties and their environments can be comprehensively analyzedby metagenomics research Associations between diseasesand specific bacteria have been described in previous studiesfor example type 2 diabetes [8ndash11] irritable bowel syndrome(IBS) [12ndash14] and colorectal cancer (CRC) [15ndash17]Moreoversome bacteria which are significantly associated with specificdiseases were thought to be biomarkers for construction of adisease risk prediction model [8 18]

Obesity is a major public health problem worldwideand its prevalence is rapidly increasing [19] Obesity isrelated to several disorders including type 2 diabetes [20ndash23] cardiovascular disease [24ndash26] and cancer [26ndash28]Recently obesity was shown to be associated with an alter-ation of the gut microbiota both in human [29ndash32] andanimal models [30 33 34] It was observed that a reducedproportion of the Bacteroidetes and increased proportionof the Firmicutes were associated with human obesity [3035 36] Also an increase of Actinobacteria in obese indi-viduals was reported [35] In another study amounts ofArchaea and Methanobacteriales were positively correlatedwith obesity [37] and their amounts in obesity samplesdecreased or disappeared after gastric bypass surgery Inaddition another study also mentioned that the amountsof Bifidobacterium and Ruminococcus decreased in obe-sity samples [38] However some studies indicated thatthe ratio of the proportions of Bacteroidetes and Firmi-cutesis contradictory [39] or not associated [5 40] withobesity

With different ethnicities and regions dietary habits andenvironmental factors can widely vary and there is a lackof studies focusing on Taiwanese samples Therefore hereinwe collected 81 stool samples from Taiwanese for analysis ofthe association between gut flora and obesity According toa study by Pan et al [41] Taiwan adopted body mass index(BMI) values of 24 and 27 as the cutoff points for beingoverweight and obese respectively In this study the stools of36 obese (BMI ge 27) and 45 normal persons (BMI le 24) were

Table 1 Study participant characteristics and demographics

Participant characteristicGender (number of samples)

Male 30Female 51

Age (years)Range 20sim89Mean 412

Height (cm)Range 1485sim181Mean 164

Weight (kg)Range 45sim110Mean 697

Body mass index (kgm2) (number of samples)le24 45ge27 36

Bacteroides (28)

Prevotella (20)

Escherichia (97)Phascolarctobacterium

(39)

Eubacterium (32)Megamonas (3)

Faecalibacterium (29)

Gemmiger (22)Sutterella (2)

Fusobacterium (19)Salmonella (19)

Megasphaera (13)Dialister (12)

Bifidobacterium (11)Akkermansia (1)

Others (17)All samples

Figure 1 The distribution of genera among all samples

collected and 16S rRNA sequencing was used to assess theassociation between obesity and the taxonomic compositionof the gut microbiota

2 Results

Participant metadata are summarized in Table 1 and detailedsample profiles are given in Table S1 available online in Sup-plementarymaterial at httpdxdoiorg1011552014906168including the number of reads gender age height weightand BMI In total 4152740 sequence reads were obtainedfrom the 81 samples and amean of 51268 readswith amedianread length of 125 bp was obtained per study participantSequence reads were processed through our taxonomic map-ping process and the distribution of genera in samples isdepicted in Figure 1 The sequencing results showed that themost abundant genera in all samples were Bacteroides (28)Prevotella (20) Escherichia (97) Phascolarctobacterium(39) Eubacterium (32) Megamonas (3) Faecalibac-terium (29) Gemmiger (22) and Sutterella (2)

BioMed Research International 3

21 Unsupervised Clustering Analysis Hierarchical clusteringwas performed using the UniFrac unweighted distance andgut bacterial communities and clinical values of each sampleare shown in Figure 2 The results demonstrate that thebacterial communities in the gutwere clustered into twomaingroups anN-like group (including theN1 andN2 subgroups)and an OB-like group (including the OB1 OB2 OB3 andOB4 subgroups) Figure 3 shows that the most abundantgenera in N-like samples were Bacteroides (278) Prevotella(186) Escherichia (127) Phascolarctobacterium (4)and Eubacterium (35) The most abundant genera in OB-like samples were Bacteroides (288) Prevotella (217)Escherichia (71) Megamonas (44) and Phascolarctobac-terium (37) Remarkably most normal samples (78) wereclustered in the N-like group and most case samples (81)were clustered in theOB-like group (Fisherrsquos119875 value = 161119864minus07) The results showed that gut bacterial community typeswere highly associated with obesity The genera diversityanalysis showed that the bacterial communities in the N-likegroup exhibited significantly higher alpha diversity and lowerbeta diversity than those in the OB-like group (Figure 4)

22 Supervised Clustering Analysis To investigate the asso-ciation between gut bacterial communities and obesity 45stool samples of participants with a BMI of le 24 were definedas normal samples and 36 samples of participants with aBMI ge 27 were used as case samples Figure 5 shows thatthe most abundant bacteria in normal samples were Bac-teroides (277) Prevotella (194) Escherichia (12) Phas-colarctobacterium (39) and Eubacterium (35) The mostabundant bacteria in case samples were Bacteroides (29)Prevotella (21) Escherichia (74) Megamonas (51) andPhascolarctobacterium (38) Normal samples had a signif-icantly higher proportion of Escherichia while case sampleshad a higher proportion ofMegamonas

Genera with significantly different proportions betweennormal and case samples are listed in Table 2 Additionallygenera with a significantly different presence between normaland case samples are listed in Table 3 and significantlydifferent species are also provided in Table S2 The generaof Shewanella Citrobacter Cronobacter Leclercia Tatumellaand Acinetobacter exhibited significant differences in bothproportions and presence Unweighted alpha and beta diver-sities of genera in the normal and case samples are shownin Figures 6(a) and 6(b) respectively The results showedthat the bacterial communities in normal samples exhibitedsignificantly higher alpha diversity and lower beta diversitythan those in casesamples

A PCoA of gut bacterial communities is shown inFigure 7 The results showed that most normal samples(green nodes) were located in the bottom left area and casesamples (red nodes) were spread in other areas (Figure 7(a))Samples in the N1 N2 OB1 OB2 OB3 and OB4 subgroupsare depicted in Figure 7(b) The results show that bacterialcommunities of N1 and N2 were highly associated withnormal-weight individuals and others were associated withobese individuals

BMI

Subgroups

N1

N2

OB1

OB2

OB3

OB4

BMI ge 27 (case)BMI le 24 (normal)

Figure 2 Bacterial communities in the samples

0

10

20

30

40

50

60

70

80

90

100

N-like OB-like

OthersPseudomonasDialisterMegasphaeraClostridiumFusobacteriumSalmonellaSutterella

GemmigerMegamonasFaecalibacteriumEubacteriumPhascolarctobacteriumEscherichiaPrevotellaBacteroides

()

Figure 3 Relatively abundant genera in the N-like and OB-likegroups


Table 2 Genera with significantly different proportions between normal and case samples

Genus Fold change(casecontrol)

KS test119875 value

ANOVA119875 value Case mean Control mean

Collinsella 056 001 037 0001 0002Barnesiella 061 001 026 0002 0003Clostridium 052 001 001 0009 0017Coprococcus 054 004 037 0001 0001Lachnospira 168 002 027 0003 0002Oscillibacter 055 001 002 0001 0001Megamonas 570 008 001 0051 0009Veillonella 047 001 027 0002 0004Shewanella 469 000 004 0001 lt0001Citrobacter 026 000 011 0002 0006Cronobacter 008 000 000 lt0001 0002Enterobacter 038 000 007 0001 0002Erwinia 018 000 000 lt0001 0002Escherichia 062 000 004 0074 0120Leclercia 009 000 006 0001 0005Morganella 005 001 007 lt0001 0001Serratia 002 000 000 lt0001 0018Tatumella 006 000 000 lt0001 0001Halomonas 351 000 008 0005 0002Acinetobacter 11608 000 003 0002 lt0001KS Kolmogorov-Smirnov ANOVA analysis of variance

Table 3 Genera with a significantly different presence between normal and case samples

Genus PresenceCasenormal

AbsenceCasenormal

Fisherrsquos test119875 value

Odds ratio(95 CI)

Butyricimonas 2844 81 0009 0082 (0002sim0664)Butyrivibrio 011 3634 0001 0088 (0002sim0663)Lachnobacterium 016 3629 lt0001 0052 (0001sim0371)Lachnospira 2243 142 lt0001 0076 (0008sim0373)Syntrophococcus 010 3635 0002 0099 (0002sim0764)Pectinatus 014 3631 lt0001 0063 (0001sim0460)Comamonas 010 3635 0002 0099 (0002sim0764)Pseudoalteromonas 121 2444 lt0001 21261 (2837sim956295)Shewanella 153 2142 lt0001 9703 (2381sim58040)Citrobacter 1743 192 lt0001 0043 (0004sim0210)Cronobacter 634 3011 lt0001 0068 (0018sim0218)Leclercia 836 289 lt0001 0075 (0021sim0233)Rahnella 013 3632 lt0001 0070 (0002sim0516)Shigella 531 3114 lt0001 0076 (0019sim0250)Tatumella 223 3422 lt0001 0058 (0006sim0273)Marinomonas 121 2444 lt0001 21261 (2837sim956295)Acinetobacter 122 2443 0001 10443 (2070sim103809)Aliivibrio 81 2844 0009 12231 (1505sim568466)CI confidence interval


OB-likeN-like

CHAO

80

70

60

50

40

30 P value = 0023

(a)

OB-likeN-like

Beta

div

ersit

y

3000

2500

2000

1500

1000

500

0P value = 0001

(b)

Figure 4 Unweighted (a) alpha diversity and (b) beta diversity of bacterial communities in the N-like and OB-like groups

Bacteroides (277)

Prevotella (194)Escherichia (12)

Phascolarctobacterium (39)


Faecalibacterium (25)Gemmiger (19)Sutterella (18)




Others (196)

Normal samples

(a)

Bacteroides (29)








Others (137)Case samples

(b)

Figure 5 Relatively abundant genera in the normal and case samples

23 Potential Markers for Classification of Normal Weightand Obesity The identified bacteria with statistical signifi-cance were used for rule-based clustering Threefold cross-validation was used to evaluate the performance of the clas-sification model Two out of the significant species in TableS2 Parabacteroides distasonis and Serratia sp DAP4 wereselected as discriminating factors in the J48 decision tree Asshown in Figure 8 the classification rules are described asfollows (1) a sample is classified as normal if Parabacteroidesdistasonis was absent (2) A sample with the presence ofParabacteroides distasonis and absence of Serratia sp DAP4was classified as a case otherwise it was classified as normalAs shown in Table S3 the classifier performed well and thearea under the receiver operating characteristic curve (AUC)

was 0813 The results showed that Parabacteroides distasonisand Serratia spDAP4might be potential markers for furtherclinical analysis and investigation of obesity

3 Discussion

Several relatively abundant genera were identified in sam-ples (Figure 5) including Bacteroides Prevotella EscherichiaPhascolarctobacterium Eubacterium Megamonas Faecal-ibacteriumGemmiger Sutterella Fusobacterium SalmonellaMegasphaera Dialister Bifidobacterium and AkkermansiaIn previous studies the presence of Bacteroides Prevotellaand Sutterella was negatively associated with obesity [36 42


CHAO

80

70

60

50

40

30 P value = 0002

CaseNormal

(a)

Beta

div

ersit

y

3000

2500

2000

1500

1000

500

0P value = 0001

CaseNormal

(b)

Figure 6 Unweighted (a) alpha diversity and (b) beta diversity of bacterial communities in case and control samples

CaseNormal

020100minus01minus02

04

03

02

01

00

minus01

minus02

PC2

(13

)

PC1 (173)

(a)

d = 01

OB-1

OB-4

N-2NOB-3OB-2

(b)

Figure 7 Unweighted principal coordinate analysis plot of (a) case and normal samples (b) samples in N1 N2 OB1 OB2 OB3 and OB4subgroups

43] In other related gastrointestinal diseases Prevotella wasincreased in children diagnosed with IBS [44] BacteroidesEubacterium and Prevotella were increased and Faecalibac-terium was reduced in CRC patients [45 46] IncreasedBacteroides and reduced Eubacterium and Prevotella werealso found in a rat model of CRC [47]

At the genus level the presence of Acinetobacter Ali-ivibrio Marinomonas Pseudoalteromonas and Shewanella

had positive associations with obesity (Table 3) Acinetobac-ter is a genus of Gram-negative bacteria and the speciesAcinetobacter baumannii is a key pathogen of infections inhospitals [48] Aliivibrio is a reclassified genus from theldquoVibrio fischeri species grouprdquo [49] and species of Aliivibrioare symbiotic with marine animals or are described as fishpathogens [50ndash52] Shewanella is a genus of marine bacteriaand some species can cause infections [53 54] Lachnospira


The presence of Parabacteroidesdistasonis in a sample

The presence of Serratia sp DAP4 in a sampleNormal

No

Case

Yes

Yes

No

Normal

Figure 8 Classification rule and potential markers for discriminat-ing between obese and normal-weight individuals

[37] Citrobacter [43] and Shigella [43] were reported tobe positively associated with obesity Lachnobacterium [18]showed a negative association with obesity

At the species level the presence of Parabacteroides dis-tasonis Lactobacillus kunkeei Pseudoalteromonas piscicidaShewanella algae Marinomonas posidonica and Aliivibriofischeri was positively associated with obesity (Table S2)Species of Bacteroides and Parabacteroides represent oppor-tunistic pathogens in infectious diseases and they are able todevelop antimicrobial drug resistance [55] Parabacteroidesdistasonis previously known as Bacteroides distasonis [56]is prominently found in the gut of healthy individuals[57] It is also related to improved human bowel healthrelease [58] and negatively associated with celiac disease[59] Our results revealed a positive association betweenParabacteroides distasonis and obesity Blautia producta [60]and Enterobacter cloacae [61] were suggested to be related toa high-fat diet causing obesity in a mouse model Serratia is agenus of Gram-negative facultatively anaerobic rod-shapedbacteria In hospitals Serratia species tend to colonize therespiratory and urinary tracts causing nosocomial infections[62 63] In related studies of the GI tract Serratia increasedin formula-fed mice [64] and was positively correlated withinfants with colic [65]

In the alpha diversity analysis (Figure 6(a)) the Chaorichness index between normal and case groups exhibited asignificant difference (119875 = 0002) This shows that bacterialcommunities in normal samples had a greater genera richnessthan those in case samples Results of the beta diversityanalysis (Figure 6(b)) showed that bacterial communities innormal samplesweremore similar than those in case samplesThe unweighted PCoA plot (Figure 7) showed that bacterialcommunities in the N-like group (including N1 and N2)were highly associated with normal individuals and bacterialcommunities in the OB-like group (including OB1 OB2OB3 and OB4) were more associated with obese individualsThe unsupervised clustering heatmap of all samples (Figure

S1(A)) was generated using Spearman correlations and theresults showed that most normal samples and most casesampleswere respectively clustered together when all generawere used for clustering However when only relativelyabundant genera were used for clustering normal and casesamples were interwoven with each other (Figure S1(B))This indicates that some genera found in small proportionsmight be important for distinguishing obese from normalindividuals

4 Conclusions

This is the first study in Taiwan to investigate the associationbetween human gut microbiota and obesity using metage-nomic sequencing The results showed that bacterial com-munities in the gut were clustered into N-like and OB-likegroups which were highly associated with normal and obesesubjects respectively Several relatively abundant bacteriawith significantly different distributions between normal andcase samples were identified and used to establish a rule-based classificationmodel Althoughdetailed functional rolesor mechanisms of these bacteria are needed for furthervalidation the results provide new insights about bacterialcommunities in the gut with a rising trend of obesity

5 Methods

51 Sample Collection and DNA Extraction Eighty-one stoolsamples were collected by Sigma-transwab (Medical Wire)into a tube with Liquid Amies Transport Medium and storedat 4∘C until being processed Fresh faeces were obtainedfrom participants and DNA was directly extracted fromstool samples using a QIAamp DNA Stool Mini Kit (Qia-gen) A swab was vigorously vortexed and incubated atroom temperature for 1min The sample was transferred tomicrocentrifuge tubes containing 560 120583L of Buffer ASL thenvortexed and incubated at 37∘C for 30min In addition thesuspension was incubated at 95∘C for 15min vortexed andcentrifuged at 14000 rpm for 1min to obtain pelletized stoolparticles Extraction was performed following the protocolof the QIAamp DNA Stool Mini Kit DNA was eluted with50 120583L Buffer AE centrifuged at 14000 rpm for 1min andthen the DNA extract was stored at minus20∘C until being furtheranalyzed

52 Library Construction and Sequencing of the V4 Region of16S rDNA The PCR primers F515 (51015840-GTGCCAGCMGCC-GCGGTAA-31015840) and R806 (51015840-GGACTACHVGGGTWTCT-AAT-31015840) were designed to amplify the V4 region of bacterial16S rDNA as described previously [66] Polymerase chainreaction (PCR) amplification was performed in a 50 120583L reac-tion volume containing 25 120583L 2x Taq Master Mix (ThermoScientific) 02 120583M of each forward and reverse primer and20 ng of a DNA template The reaction conditions includedan initial temperature of 95∘C for 5min followed by 30cycles of 95∘C for 30 s 54∘C for 1min and 72∘C for 1minwith a final extension of 72∘C for 5min Next amplifiedproducts were checked by 2 agarose gel electrophoresis


and ethidium bromide staining Amplicons were purifiedusing the AMPure XP PCR Purification Kit (Agencourt) andquantified using the Qubit dsDNA HS Assay Kit (Qubit)on a Qubit 20 Fluorometer (Qubit) all according to therespective manufacturerrsquos instructions For V4 library prepa-ration Illumina adapters were attached to the ampliconsusing the Illumina TruSeq DNA Sample Preparation v2 KitPurified libraries were processed for cluster generation andsequencing using the MiSeq system

53 Filtering 16S rRNA (rDNA) Sequencing Data for QualityTheFASTX-Toolkit (httphannonlabcshledufastx toolkit)was used to process the raw fastq read data files from IlluminaMiseq The sequence quality criteria were as follows (1) theminimum acceptable phred quality score of sequences was 30with a score ofgt70of sequence bases ofge20 (2) after qualitytrimming from the sequence tail sequences of gt100 bp wereretained and they also had an acceptable phred quality scoreof 30 and (3) both forward and reverse sequencing readswhich met the first and second requirements were retainedfor subsequent analysis Sequencing reads from differentsamples were identified and separated according to specificbarcodes in the 51015840 end of the sequence (with two mismatchesallowed)

54 Taxonomic Assignments of Bacterial 16S rRNA SequencesPaired-end sequences were obtained and their qualities wereassessed using the FASTX-Toolkit To generate taxonomicassignments Bowtie2 was used to align sequencing readsagainst the collection of a 16S rRNA sequences database Astandard of 97 similarity against the database was applied16S rRNA sequences of bacteria were retrieved from theSILVA ribosomal RNA sequence database [67] Followingsequence data collection sequences were extracted using V4forward and reverse primers To prevent repetitive sequenceassignments V4 sequences from SILVA were then clusteredinto several clusters by 97 similarity using UCLUST [68]Results of the taxonomic assignment were filtered to retainassignments with gt10 sequences

55 Bacterial Community Analysis After taxonomic assign-ment an operational taxonomic unit (OTU) table wasgenerated To normalize the sample size of all samples ararefaction process was performed on the OTU table Alphaand beta diversities were calculated based on a rarifiedOTU table The Kolmogorov-Smirnov test and an analysis ofvariance (ANOVA) test with the Bonferroni correction wereused to investigate significant differences between differentsample groups To observe relationships between samples andexplore taxonomic associations weighted and unweightedUniFrac [69] distance metrics were also generated basedon the rarified OTU table A principal coordinate analysis(PCoA) and unsupervised clustering were performed basedon the UniFac distance matrix To explore relationshipsbetween clinical features and different sample groups Spear-manrsquos correlation coefficient and regression analysis wereperformed The statistical analytical process was done inR language The J48 machine learning method in Weka

367 [70] was used to construct a classification rule fordiscriminating between obese and normal individuals

Conflict of Interests

The authors declare that there is no conflict of interestsregarding the publication of this paper

Authorsrsquo Contribution

Chih-Min Chiu Wei-Chih Huang and Shun-Long Wengcontributed equally to this work

Acknowledgments

The authors would like to thank the National Science Councilof the Republic of China for financially supporting thisresearch under Grant nos NSC 101-2311-B-009-005-MY3 andNSC 102-2627-B-009-001 This work was supported in partby the Ministry of Science and Technology under Grant noMOST 103-2221-E-038-013-MY2 This work was supportedin part by UST-UCSD International Center of Excellencein Advanced Bioengineering sponsored by the Ministry ofScience and Technology I-RiCE Program under Grant noNSC-102-2911-I-009-101- and Veterans General Hospitalsand University System of Taiwan (VGHUST) Joint ResearchProgram under Grant no VGHUST103-G5-1-2 This workwas also partially supported by MOE ATU

References

[1] J R Goldsmith and R B Sartor ldquoThe role of diet on intestinalmicrobiota metabolism downstream impacts on host immunefunction and health and therapeutic implicationsrdquo Journal ofGastroenterology vol 49 no 5 pp 785ndash798 2014

[2] J Chen X He and J Huang ldquoDiet effects in gut microbiomeand obesityrdquo Journal of Food Science vol 79 no 4 pp R442ndashR451 2014

[3] O O Erejuwa S A Sulaiman and M S Wahab ldquoModulationof gut microbiota in the management of metabolic disordersthe prospects and challengesrdquo International Journal ofMolecularSciences vol 15 no 3 pp 4158ndash4188 2014

[4] G L HoldM Smith C Grange E RWatt EM El-Omar andI Mukhopadhya ldquoRole of the gut microbiota in inflammatorybowel disease pathogenesis what have we learnt in the past 10yearsrdquo World Journal of Gastroenterology vol 20 no 5 pp1192ndash1210 2014

[5] M Arumugam J Raes E Pelletier et al ldquoEnterotypes of thehuman gutmicrobiomerdquoNature vol 473 no 7346 pp 174ndash1802011

[6] G D Wu J Chen C Hoffmann et al ldquoLinking long-termdietary patterns with gut microbial enterotypesrdquo Science vol334 no 6052 pp 105ndash108 2011

[7] J Peterson S Garges M Giovanni et al ldquoThe NIH humanmicrobiome projectrdquoGenome Research vol 19 no 12 pp 2317ndash2323 2009

[8] J Qin Y Li Z Cai et al ldquoAmetagenome-wide association studyof gut microbiota in type 2 diabetesrdquo Nature vol 490 no 7418pp 55ndash60 2012


[9] F H Karlsson V Tremaroli I Nookaew et al ldquoGut metage-nome in European women with normal impaired and diabeticglucose controlrdquo Nature vol 498 no 7452 pp 99ndash103 2013

[10] X Zhang D Shen Z Fang et al ldquoHuman gut microbiotachanges reveal the progression of glucose intolerancerdquo PLoSONE vol 8 no 8 Article ID e71108 2013

[11] F Karlsson V Tremaroli J Nielsen and F Backhed ldquoAssessingthe human gut microbiota in metabolic diseasesrdquo Diabetes vol62 no 10 pp 3341ndash3349 2013

[12] A Lyra T Rinttila J Nikkila et al ldquoDiarrhoea-predominantirritable bowel syndrome distinguishable by 16S rRNA genephylotype quantificationrdquo World Journal of Gastroenterologyvol 15 no 47 pp 5936ndash5945 2009

[13] S O Noor K Ridgway L Scovell et al ldquoUlcerative colitis andirritable bowel patients exhibit distinct abnormalities of the gutmicrobiotardquo BMC Gastroenterology vol 10 article 134 2010

[14] M Rajilic-Stojanovic E Biagi H G H J Heilig et al ldquoGlobaland deep molecular analysis of microbiota signatures in fecalsamples from patients with irritable bowel syndromerdquo Gas-troenterology vol 141 no 5 pp 1792ndash1801 2011

[15] Y Zhu T Michelle Luo C Jobin and H A Young ldquoGutmicrobiota and probiotics in colon tumorigenesisrdquo CancerLetters vol 309 no 2 pp 119ndash127 2011

[16] M A Hullar A N Burnett-Hartman and J W Lampe ldquoGutmicrobes diet and cancerrdquoCancer Treatment and Research vol159 pp 377ndash399 2014

[17] H R Hagland and K Soreide ldquoCellular metabolism in colorec-tal carcinogenesis influence of lifestyle gut microbiome andmetabolic pathwaysrdquo Cancer Letters 2014

[18] K Korpela H J Flint A M Johnstone et al ldquoGut microbiotasignatures predict host and microbiota responses to dietaryinterventions in obese individualsrdquo PLoS One vol 9 no 3Article ID e90702 2014

[19] B A Swinburn G Sacks K D Hall et al ldquoThe global obesitypandemic shaped by global drivers and local environmentsrdquoThe Lancet vol 378 no 9793 pp 804ndash814 2011

[20] K C Portero McLellan K Wyne E T Villagomez and WA Hsueh ldquoTherapeutic interventions to reduce the risk ofprogression from prediabetes to type 2 diabetes mellitusrdquoTherapeutics and Clinical RiskManagement vol 10 pp 173ndash1882014

[21] A Everard and P D Cani ldquoDiabetes obesity and gut micro-biotardquo Best Practice and Research Clinical Gastroenterology vol27 no 1 pp 73ndash83 2013

[22] J A Bell M Kivimaki and M Hamer ldquoMetabolically healthyobesity and risk of incident type 2 diabetes a meta-analysis ofprospective cohort studiesrdquo Obesity Reviews vol 15 no 6 pp504ndash515 2014

[23] D Nagakubo M Shirai Y Nakamura et al ldquoProphylacticeffects of the glucagon-like Peptide-1 analog liraglutide onhyperglycemia in a rat model of type 2 diabetes mellitusassociated with chronic pancreatitis and obesityrdquo ComparativeMedicine vol 64 no 2 pp 121ndash127 2014

[24] M Kratz T Baars and S Guyenet ldquoThe relationship betweenhigh-fat dairy consumption and obesity cardiovascular andmetabolic diseaserdquo European Journal of Nutrition vol 52 no1 pp 1ndash24 2013

[25] G M Hinnouho S Czernichow A Dugravot et al ldquoMetaboli-cally healthy obesity and the risk of cardiovascular disease andtype 2 diabetesTheWhitehall II Cohort Studyrdquo EuropeanHeartJournal 2014

[26] K A Britton J M Massaro J M Murabito B E KregerU Hoffmann and C S Fox ldquoBody fat distribution incidentcardiovascular disease cancer and all-cause mortalityrdquo Journalof the American College of Cardiology vol 62 no 10 pp 921ndash925 2013

[27] E E FrezzaM SWachtel andMChiriva-Internati ldquoInfluenceof obesity on the risk of developing colon cancerrdquo Gut vol 55no 2 pp 285ndash291 2006

[28] K Nakamura A Hongo J Kodama and Y Hiramatsu ldquoFataccumulation in adipose tissues as a risk factor for the devel-opment of endometrial cancerrdquoOncology Reports vol 26 no 1pp 65ndash71 2011

[29] M Remely E Aumueller D Jahn B Hippe H Brath and AG Haslberger ldquoMicrobiota and epigenetic regulation of inflam-matory mediators in type 2 diabetes and obesityrdquo BeneficialMicrobes vol 5 no 1 pp 33ndash43 2014

[30] R E Ley P J Turnbaugh S Klein and J I Gordon ldquoMicrobialecology human gut microbes associated with obesityrdquo Naturevol 444 no 7122 pp 1022ndash1023 2006

[31] J Shen M S Obin and L Zhao ldquoThe gut microbiota obesityand insulin resistancerdquo Molecular Aspects of Medicine vol 34no 1 pp 39ndash58 2013

[32] M Nieuwdorp P W Gilijamse N Pai and L M KaplanldquoRole of the microbiome in energy regulation andmetabolismrdquoGastroenterology vol 146 no 6 pp 1525ndash1533 2014

[33] P J Turnbaugh F Backhed L Fulton and J I Gordon ldquoDiet-induced obesity is linked to marked but reversible alterations inthe mouse distal gut microbiomerdquo Cell Host andMicrobe vol 3no 4 pp 213ndash223 2008

[34] F Backhed H Ding T Wang et al ldquoThe gut microbiota as anenvironmental factor that regulates fat storagerdquo Proceedings ofthe National Academy of Sciences of the United States of Americavol 101 no 44 pp 15718ndash15723 2004

[35] P J Turnbaugh M Hamady T Yatsunenko et al ldquoA core gutmicrobiome in obese and lean twinsrdquoNature vol 457 no 7228pp 480ndash484 2009

[36] J Furet L Kong J Tap et al ldquoDifferential adaptation of humangut microbiota to bariatric surgery-induced weight loss linkswithmetabolic and low-grade inflammationmarkersrdquoDiabetesvol 59 no 12 pp 3049ndash3057 2010

[37] H Zhang J K DiBaise A Zuccolo et al ldquoHuman gutmicrobiota in obesity and after gastric bypassrdquo Proceedings ofthe National Academy of Sciences of the United States of Americavol 106 no 7 pp 2365ndash2370 2009

[38] Y N Yin Q F Yu N Fu XW Liu and F G Lu ldquoEffects of fourBifidobacteria on obesity in high-fat diet induced ratsrdquo WorldJournal of Gastroenterology vol 16 no 27 pp 3394ndash3401 2010

[39] R E Ley ldquoObesity and the human microbiomerdquo CurrentOpinion in Gastroenterology vol 26 no 1 pp 5ndash11 2010

[40] TheHumanMicrobiome Project Consortium ldquoStructure func-tion and diversity of the healthy human microbiomerdquo Naturevol 486 no 7402 pp 207ndash214 2012

[41] W Pan K M Flegal H Chang W Yeh C Yeh and WLee ldquoBody mass index and obesity-related metabolic disordersin Taiwanese and US whites and blacks Implications fordefinitions of overweight and obesity for AsiansrdquoTheAmericanJournal of Clinical Nutrition vol 79 no 1 pp 31ndash39 2004

[42] A Santacruz M C Collado L Garcıa-Valdes et al ldquoGutmicrobiota composition is associated with body weight weightgain and biochemical parameters in pregnant womenrdquo BritishJournal of Nutrition vol 104 no 1 pp 83ndash92 2010


[43] S Xiao N Fei X Pang et al ldquoA gut microbiota-targeteddietary intervention for amelioration of chronic inflammationunderlying metabolic syndromerdquo FEMS Microbiology Ecologyvol 87 no 2 pp 357ndash367 2014

[44] L Rigsbee R Agans V Shankar et al ldquoQuantitative profiling ofgut microbiota of children with diarrhea-predominant irritablebowel syndromerdquo The American Journal of Gastroenterologyvol 107 no 11 pp 1740ndash1751 2012

[45] N Wu X Yang R Zhang et al ldquoDysbiosis signature of fecalmicrobiota in colorectal cancer patientsrdquoMicrobial Ecology vol66 no 2 pp 462ndash470 2013

[46] W Chen F Liu Z Ling X Tong and C Xiang ldquoHumanintestinal lumen and mucosa-associated microbiota in patientswith colorectal cancerrdquo PLoS ONE vol 7 no 6 Article IDe39743 2012

[47] Q Zhu Z Jin W Wu et al ldquoAnalysis of the intestinal lumenmicrobiota in an animalmodel of colorectal cancerrdquo PLoSONEvol 9 no 3 Article ID e90849 2014

[48] DWang D Yan W Hou X Zeng Y Qi and J Chen ldquoCharac-terization of bla

119874119909119860minus23

gene regions in isolates of Acinetobacterbaumanniirdquo Journal of Microbiology Immunology and Infection2014

[49] H Urbanczyk J C Ast M J Higgins J Carson and P VDunlap ldquoReclassification of Vibrio fischeri Vibrio logei Vibriosalmonicida and Vibrio wodanis as Aliivibrio fischeri gen novcomb nov Aliivibrio logei comb nov Aliivibrio salmonicidacomb nov and Aliivibrio wodanis comb novrdquo InternationalJournal of Systematic and EvolutionaryMicrobiology vol 57 part12 pp 2823ndash2829 2007

[50] J C Ast H Urbanczyk and P V Dunlap ldquoMulti-gene analysisreveals previously unrecognized phylogenetic diversity in Ali-ivibriordquo Systematic and Applied Microbiology vol 32 no 6 pp379ndash386 2009

[51] R Beaz-Hidalgo A Doce S Balboa J L Barja and J LRomalde ldquoAliivibrio finisterrensis sp nov isolated fromManilaclam Ruditapes philippinarum and emended description ofthe genus Aliivibriordquo International Journal of Systematic andEvolutionary Microbiology vol 60 no 1 pp 223ndash228 2010

[52] S Yoshizawa H Karatani M Wada A Yokota and K KogureldquoAliivibrio sifiae sp nov luminous marine bacteria isolatedfrom seawaterrdquo Journal of General and Applied Microbiologyvol 56 no 6 pp 509ndash518 2010

[53] Y S Chen Y C Liu M Y Yen J H Wang S R Wann andD L Cheng ldquoSkin and soft-tissue manifestations of Shewanellaputrefaciens infectionrdquo Clinical Infectious Diseases vol 25 no2 pp 225ndash229 1997

[54] N Vignier M Barreau C Olive et al ldquoHuman infectionwith Shewanella putrefaciens and S algae report of 16 cases inMartinique and review of the literaturerdquo American Journal ofTropical Medicine and Hygiene vol 89 no 1 pp 151ndash156 2013

[55] R F Boente L Q Ferreira L S Falcao et al ldquoDetection ofresistance genes and susceptibility patterns in Bacteroides andParabacteroides strainsrdquo Anaerobe vol 16 no 3 pp 190ndash1942010

[56] M Sakamoto and Y Benno ldquoReclassification of Bacteroidesdistasonis Bacteroides goldsteinii and Bacteroides merdae asParabacteroides distasonis gen nov comb nov Parabac-teroides goldsteinii comb nov and Parabacteroides merdaecomb novrdquo International Journal of Systematic and EvolutionaryMicrobiology vol 56 part 7 pp 1599ndash1605 2006

[57] J Xu M A Mahowald R E Ley et al ldquoEvolution of symbioticbacteria in the distal human intestinerdquo PLoS Biology vol 5 no7 Article ID e156 2007

[58] JMClarke D L Topping C T Christophersen et al ldquoButyrateesterified to starch is released in the human gastrointestinaltractrdquo The American Journal of Clinical Nutrition vol 94 no5 pp 1276ndash1283 2011

[59] E Sanchez E Donat C Ribes-Koninckx M Calabuig andY Sanz ldquoIntestinal Bacteroides species associated with coeliacdiseaserdquo Journal of Clinical Pathology vol 63 no 12 pp 1105ndash1111 2010

[60] N Becker J Kunath G Loh and M Blaut ldquoHuman intestinalmicrobiota characterization of a simplified and stable gnotobi-otic rat modelrdquo Gut Microbes vol 2 no 1 2011

[61] N Fei and L Zhao ldquoAn opportunistic pathogen isolated fromthe gut of an obese human causes obesity in germfree micerdquoISME Journal vol 7 no 4 pp 880ndash884 2013

[62] A Hejazi and F R Falkiner ldquoSerratia marcescensrdquo Journal ofMedical Microbiology vol 46 no 11 pp 903ndash912 1997

[63] M Patankar S Sukumaran A Chhibba U Nayak andL Sequeira ldquoComparative in-vitro activity of cefoperazone-tazobactam and cefoperazone-sulbactam combinations againstESBL pathogens in respiratory and urinary infectionsrdquo Journalof Association of Physicians of India vol 60 no 11 pp 22ndash242012

[64] E M Carlisle V Poroyko M S Caplan J Alverdy M JMorowitz and D Liu ldquoMurine gut microbiota and transcrip-tome are diet dependentrdquo Annals of Surgery vol 257 no 2 pp287ndash294 2013

[65] C de Weerth S Fuentes P Puylaert and W M de vosldquoIntestinal microbiota of infants with colic development andspecific signaturesrdquo Pediatrics vol 131 no 2 pp e550ndashe5582013

[66] J G Caporaso C L Lauber W A Walters et al ldquoGlobalpatterns of 16S rRNA diversity at a depth of millions ofsequences per samplerdquo Proceedings of the National Academy ofSciences of the United States of America vol 108 no 1 pp 4516ndash4522 2011

[67] E Pruesse C Quast K Knittel et al ldquoSILVA a comprehensiveonline resource for quality checked and aligned ribosomal RNAsequence data compatible with ARBrdquo Nucleic Acids Researchvol 35 no 21 pp 7188ndash7196 2007

[68] R C Edgar ldquoSearch and clustering orders of magnitude fasterthan BLASTrdquo Bioinformatics vol 26 no 19 pp 2460ndash24612010

[69] C Lozupone M E Lladser D Knights J Stombaugh and RKnight ldquoUniFrac an effective distance metric for microbialcommunity comparisonrdquo ISME Journal vol 5 no 2 pp 169ndash172 2011

[70] M Hall E Frank G Holmes B Pfahringer P Reutemann andI H Witten ldquoThe WEKA data mining software an updaterdquoSIGKDD Explorations vol 11 no 1 pp 10ndash18 2009

Submit your manuscripts athttpwwwhindawicom

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Anatomy Research International

PeptidesInternational Journal of


Hindawi Publishing Corporation httpwwwhindawicom

International Journal of

Volume 2014

Zoology


Molecular Biology International

GenomicsInternational Journal of


The Scientific World JournalHindawi Publishing Corporation httpwwwhindawicom Volume 2014


BioinformaticsAdvances in

Marine BiologyJournal of



Signal TransductionJournal of


BioMed Research International

Evolutionary BiologyInternational Journal of



Biochemistry Research International

ArchaeaHindawi Publishing Corporationhttpwwwhindawicom Volume 2014


Genetics Research International


Advances in

Virolog y

Hindawi Publishing Corporationhttpwwwhindawicom

Nucleic AcidsJournal of

Volume 2014

Stem CellsInternational



Enzyme Research



Microbiology


were associated with dietary habits [6] Animal protein andsaturated fats were highly correlated with the Bacteroidesenterotype Low meat intake and plant-based nutritionwith high carbohydrates were correlated with the Prevotellaenterotype

TheHumanMicrobiomeProject (HMP) [7]was launchedby the US National Institutes of Health in 2008 andunderstanding the relationship between human health andmicrobiota that live in or on the human body was recognizedas an important concept To decipher this relationship high-throughput sequencing also called next-generation sequenc-ing (NGS) supporting a large number of sequences can beused to sequence 16S ribosomal (r)RNA to construct complexmicrobial community profiles 16S rRNA is considered thestandard for studying microbial communities and assigningtaxonomy to bacteria Compared to conventional polymerasechain reaction- (PCR-) based or culture-based methods 16SrRNA sequencing by NGS can detect hundreds to thousandsof bacteria at one time and offer relative quantification of thebacteria Interactions between different bacterial communi-ties and their environments can be comprehensively analyzedby metagenomics research Associations between diseasesand specific bacteria have been described in previous studiesfor example type 2 diabetes [8ndash11] irritable bowel syndrome(IBS) [12ndash14] and colorectal cancer (CRC) [15ndash17]Moreoversome bacteria which are significantly associated with specificdiseases were thought to be biomarkers for construction of adisease risk prediction model [8 18]

Obesity is a major public health problem worldwideand its prevalence is rapidly increasing [19] Obesity isrelated to several disorders including type 2 diabetes [20ndash23] cardiovascular disease [24ndash26] and cancer [26ndash28]Recently obesity was shown to be associated with an alter-ation of the gut microbiota both in human [29ndash32] andanimal models [30 33 34] It was observed that a reducedproportion of the Bacteroidetes and increased proportionof the Firmicutes were associated with human obesity [3035 36] Also an increase of Actinobacteria in obese indi-viduals was reported [35] In another study amounts ofArchaea and Methanobacteriales were positively correlatedwith obesity [37] and their amounts in obesity samplesdecreased or disappeared after gastric bypass surgery Inaddition another study also mentioned that the amountsof Bifidobacterium and Ruminococcus decreased in obe-sity samples [38] However some studies indicated thatthe ratio of the proportions of Bacteroidetes and Firmi-cutesis contradictory [39] or not associated [5 40] withobesity

With different ethnicities and regions dietary habits andenvironmental factors can widely vary and there is a lackof studies focusing on Taiwanese samples Therefore hereinwe collected 81 stool samples from Taiwanese for analysis ofthe association between gut flora and obesity According toa study by Pan et al [41] Taiwan adopted body mass index(BMI) values of 24 and 27 as the cutoff points for beingoverweight and obese respectively In this study the stools of36 obese (BMI ge 27) and 45 normal persons (BMI le 24) were

Table 1 Study participant characteristics and demographics

Participant characteristicGender (number of samples)

Male 30Female 51

Age (years)Range 20sim89Mean 412

Height (cm)Range 1485sim181Mean 164

Weight (kg)Range 45sim110Mean 697

Body mass index (kgm2) (number of samples)le24 45ge27 36

Bacteroides (28)

Prevotella (20)

Escherichia (97)Phascolarctobacterium

(39)


Faecalibacterium (29)

Gemmiger (22)Sutterella (2)




Others (17)All samples

Figure 1 The distribution of genera among all samples

collected and 16S rRNA sequencing was used to assess theassociation between obesity and the taxonomic compositionof the gut microbiota

2 Results

Participant metadata are summarized in Table 1 and detailedsample profiles are given in Table S1 available online in Sup-plementarymaterial at httpdxdoiorg1011552014906168including the number of reads gender age height weightand BMI In total 4152740 sequence reads were obtainedfrom the 81 samples and amean of 51268 readswith amedianread length of 125 bp was obtained per study participantSequence reads were processed through our taxonomic map-ping process and the distribution of genera in samples isdepicted in Figure 1 The sequencing results showed that themost abundant genera in all samples were Bacteroides (28)Prevotella (20) Escherichia (97) Phascolarctobacterium(39) Eubacterium (32) Megamonas (3) Faecalibac-terium (29) Gemmiger (22) and Sutterella (2)






BMI

Subgroups

N1

N2

OB1

OB2

OB3

OB4



0

10

20

30

40

50

60

70

80

90

100

N-like OB-like



()





KS test119875 value





AbsenceCasenormal


Odds ratio(95 CI)



OB-likeN-like

CHAO

80

70

60

50

40

30 P value = 0023

(a)

OB-likeN-like

Beta

div

ersit

y

3000

2500

2000

1500

1000

500

0P value = 0001

(b)


Bacteroides (277)








Others (196)

Normal samples

(a)

Bacteroides (29)









(b)




3 Discussion



CHAO

80

70

60

50

40

30 P value = 0002

CaseNormal

(a)

Beta

div

ersit

y

3000

2500

2000

1500

1000

500

0P value = 0001

CaseNormal

(b)


CaseNormal


04

03

02

01

00

minus01

minus02

PC2

(13

)

PC1 (173)

(a)

d = 01

OB-1

OB-4

N-2NOB-3OB-2

(b)








No

Case

Yes

Yes

No

Normal






4 Conclusions


5 Methods













Acknowledgments


References



















































119874119909119860minus23































Volume 2014

Zoology






















Advances in

Virolog y



Volume 2014




Enzyme Research



Microbiology






BMI

Subgroups

N1

N2

OB1

OB2

OB3

OB4



0

10

20

30

40

50

60

70

80

90

100

N-like OB-like



()





KS test119875 value





AbsenceCasenormal


Odds ratio(95 CI)



OB-likeN-like

CHAO

80

70

60

50

40

30 P value = 0023

(a)

OB-likeN-like

Beta

div

ersit

y

3000

2500

2000

1500

1000

500

0P value = 0001

(b)


Bacteroides (277)








Others (196)

Normal samples

(a)

Bacteroides (29)









(b)




3 Discussion



CHAO

80

70

60

50

40

30 P value = 0002

CaseNormal

(a)

Beta

div

ersit

y

3000

2500

2000

1500

1000

500

0P value = 0001

CaseNormal

(b)


CaseNormal


04

03

02

01

00

minus01

minus02

PC2

(13

)

PC1 (173)

(a)

d = 01

OB-1

OB-4

N-2NOB-3OB-2

(b)








No

Case

Yes

Yes

No

Normal






4 Conclusions


5 Methods













Acknowledgments


References



















































119874119909119860minus23































Volume 2014

Zoology






















Advances in

Virolog y



Volume 2014




Enzyme Research



Microbiology




KS test119875 value





AbsenceCasenormal


Odds ratio(95 CI)



OB-likeN-like

CHAO

80

70

60

50

40

30 P value = 0023

(a)

OB-likeN-like

Beta

div

ersit

y

3000

2500

2000

1500

1000

500

0P value = 0001

(b)


Bacteroides (277)








Others (196)

Normal samples

(a)

Bacteroides (29)









(b)




3 Discussion



CHAO

80

70

60

50

40

30 P value = 0002

CaseNormal

(a)

Beta

div

ersit

y

3000

2500

2000

1500

1000

500

0P value = 0001

CaseNormal

(b)


CaseNormal


04

03

02

01

00

minus01

minus02

PC2

(13

)

PC1 (173)

(a)

d = 01

OB-1

OB-4

N-2NOB-3OB-2

(b)








No

Case

Yes

Yes

No

Normal






4 Conclusions


5 Methods













Acknowledgments


References



















































119874119909119860minus23































Volume 2014

Zoology






















Advances in

Virolog y



Volume 2014




Enzyme Research



Microbiology


OB-likeN-like

CHAO

80

70

60

50

40

30 P value = 0023

(a)

OB-likeN-like

Beta

div

ersit

y

3000

2500

2000

1500

1000

500

0P value = 0001

(b)


Bacteroides (277)








Others (196)

Normal samples

(a)

Bacteroides (29)









(b)




3 Discussion



CHAO

80

70

60

50

40

30 P value = 0002

CaseNormal

(a)

Beta

div

ersit

y

3000

2500

2000

1500

1000

500

0P value = 0001

CaseNormal

(b)


CaseNormal


04

03

02

01

00

minus01

minus02

PC2

(13

)

PC1 (173)

(a)

d = 01

OB-1

OB-4

N-2NOB-3OB-2

(b)








No

Case

Yes

Yes

No

Normal






4 Conclusions


5 Methods













Acknowledgments


References



















































119874119909119860minus23































Volume 2014

Zoology






















Advances in

Virolog y



Volume 2014




Enzyme Research



Microbiology


CHAO

80

70

60

50

40

30 P value = 0002

CaseNormal

(a)

Beta

div

ersit

y

3000

2500

2000

1500

1000

500

0P value = 0001

CaseNormal

(b)


CaseNormal


04

03

02

01

00

minus01

minus02

PC2

(13

)

PC1 (173)

(a)

d = 01

OB-1

OB-4

N-2NOB-3OB-2

(b)








No

Case

Yes

Yes

No

Normal






4 Conclusions


5 Methods













Acknowledgments


References



















































119874119909119860minus23































Volume 2014

Zoology






















Advances in

Virolog y



Volume 2014




Enzyme Research



Microbiology




No

Case

Yes

Yes

No

Normal






4 Conclusions


5 Methods













Acknowledgments


References



















































119874119909119860minus23































Volume 2014

Zoology






















Advances in

Virolog y



Volume 2014




Enzyme Research



Microbiology











Acknowledgments


References



















































119874119909119860minus23































Volume 2014

Zoology






















Advances in

Virolog y



Volume 2014




Enzyme Research



Microbiology











































119874119909119860minus23































Volume 2014

Zoology






















Advances in

Virolog y



Volume 2014




Enzyme Research



Microbiology








Volume 2014

Zoology






















Advances in

Virolog y



Volume 2014




Enzyme Research



Microbiology

research article systematic analysis of the association...

Documents