Microbial diversity and community composition of caecal microbiota in commercial and indigenous Indian chickens determined using 16s rDNA amplicon sequencing
The caecal microbiota plays a key role in chicken health and performance, influencing digestion and absorption of nutrients, and contributing to defence against colonisation by invading pathogens. Measures of productivity and resistance to pathogen colonisation are directly influenced by chicken genotype, but host driven variation in microbiome structure is also likely to exert a considerable indirect influence.
Here, we define the caecal microbiome of indigenous Indian Aseel and Kadaknath chicken breeds and compare them with the global commercial broiler Cobb400 and Ross 308 lines using 16S rDNA V3-V4 hypervariable amplicon sequencing.
Each caecal microbiome was dominated by the genera Bacteroides, unclassified bacteria, unclassified Clostridiales, Clostridium, Alistipes, Faecalibacterium, Eubacterium and Blautia. Geographic location (a measure recognised to include variation in environmental and climatic factors, but also likely to feature varied management practices) and chicken line/breed were both found to exert significant impacts (p < 0.05) on caecal microbiome composition. Linear discriminant analysis effect size (LEfSe) revealed 42 breed-specific biomarkers in the chicken lines reared under controlled conditions at two different locations.
Chicken breed-specific variation in bacterial occurrence, correlation between genera and clustering of operational taxonomic units indicate scope for quantitative genetic analysis and the possibility of selective breeding of chickens for defined enteric microbiota.
KeywordsAmplicon sequencing Chickens Microbiome Pathogens 16S rRNA gene
Ensuring the secure supply of safe food is a major global concern. Increasing human population sizes, income levels and urbanisation all contribute to rising demand for protein, and therefore livestock [1, 2]. However, key constraints on animal production include infectious diseases, which can be exacerbated by suboptimal commensal microflora that undermine capacity to thrive .
Chickens are the most numerous livestock in the world, with relatively low production costs and highly efficient food conversion. More than 60 billion chickens are produced annually, with production predicted to increase further over the next 20 years, particularly in South Asia and Africa . The importance of the caecal microbiome to chicken health and productivity has long been recognised, especially in food conversion, resistance to disease and colonisation by zoonotic pathogens [5, 6, 7, 8]. In line with other next-generation sequencing (NGS) microbiome studies [9, 10, 11, 12, 13, 14, 15], we have previously identified Bacteroidetes, Firmicutes and Actinobacteria as highly abundant phyla within chicken caecal populations, while functional metagenomics analysis revealed enrichment of sequences corresponding to carbohydrate metabolism . However, the influence of host genetics on caecal microbiome structure is unclear [17, 18, 19].
Reports of higher productivity, product quality and/or pathogen resistance have been attributed to some indigenous chicken breeds. For example, 17 breeds of Indian origin are currently registered with the National Bureau of Animal Genetic Resources (NBAGR) at the Indian Council of Agricultural Research (ICAR) (http://www.nbagr.res.in/regchi.html). Among these breeds examples such as Aseel and Kadaknath have been associated with improved egg, meat and welfare traits such as reduced feather pecking [20, 21], resistance to infectious disease [22, 23] and immune parameters distinct from those of modern commercial chickens [24, 25]. Variation in immunity-related pathways can influence microbiome structure in humans  and is also likely to affect chicken microbiomes . Analysis defining the caecal microbiome of native Indian chicken breeds of Assam state has been reported recently , although the significant impact of environmental factors makes it impossible to compare between microbiome studies . Evidence from humans, mice and chickens suggest that host genotype can exert a strong influence on microbiome composition [18, 19] and thereby offers a quantifiable phenotype which may differ between chicken breeds and individuals, and may be amenable to genetic selection. Here, we use NGS targeting hypervariable regions within microbial 16S rRNA genes to compare the caecal lumen microbiota of commercial broiler-type lines, which are comparable to those used around the world, to indigenous Indian breeds reared in parallel under controlled commercial-type management conditions. Our hypothesis was that the caecal lumen microflora would vary significantly between chicken breeds and lines, offering opportunities for targeted genetic improvement by selective breeding. Given the importance of primer choice to successful NGS [29, 30, 31, 32, 33, 34], we also compared multiple primers targeting different variable regions of the 16S rRNA.
Chicken breeds and experimental design
Four chicken breeds or lines were chosen for comparison in this study including two indigenous Indian breeds and two global commercial broiler lines. The indigenous breeds Kadaknath and Aseel were chosen for use at the Central Poultry Research Station of Anand Agricultural University, (AAU, Gujarat, India; termed location 1) and duplicated at Tamil Nadu Veterinary and Animal Sciences University (TANUVAS, Chennai, Tamil Nadu, India; location 2) (Additional file 1). Commercial Broilers of the Cobb400 (AAU) and Ross 308 (TANUVAS) lines were used for comparison as representatives of the local dominant-intensive commercial production systems .
At each location, ten chickens of each breed were hatched on the same day and reared in neighbouring pens in single poultry houses, providing independent replication and controlling against local spatial variation. Chickens were accommodated in this manner to avoid husbandry (behavioural) problems from arising when the chicken breeds were mixed. Care was taken to ensure that pens were of equal sizes, equidistant to the door and light sources and received exactly the same husbandry to minimise non-host variation. A deep litter system was employed using rice husk as a substrate in common with local practices. All chickens were fed a standard maize and soybean-based commercial diet which included bacitracin methylene disalicylate (BMD) and maduramycin (10%) for routine prophylaxis.
Sample collection and DNA extractions
At each location, five apparently healthy chickens were selected at random from each group, caught and euthanized by cervical dislocation at 42 days of age. Both caecal pouches were opened immediately using sterile scissors and the contents were recovered into sterile cryovials containing Bacterial Protect RNA reagent (QIAGEN, Germany) at an approximate 1:1 ratio (w/v). Each sample was immediately stored in a portable freezer at − 20 °C, transported to the laboratory and stored at − 80 °C.
Total genomic DNA was extracted from the pooled caecal contents of each individual chicken using the commercially available QIAamp Fast DNA Stool Mini kit (QIAGEN, Germany) following the manufacturer’s instructions with some modifications. Briefly, 300 μL of caecal content with Bacterial Protect RNA reagent was added to 1 mL of InhibitEX buffer, vortexed at 2800 rpm for 1 min to homogenise and incubated at 80 °C for 10 min. The mixture was then centrifuged at 2600 g for 30 s to remove residual solid material and 600 μL of supernatant was processed as recommended by the manufacturer. DNA was treated with DNase free RNase (Macherey-Nagel, Germany) to remove contaminating RNA. DNA concentration and quality were assessed using a Qubit 2.0 fluorometer (Invitrogen, ThermoFisher scientific, MA) and gel electrophoresis. DNA was stored at − 20 °C until further processing.
16S rRNA gene amplification and MiSeq sequencing
Six hypervariable regions within the 16S rRNA gene were amplified from caecal DNA sample using three different primer pairs (Additional file 2). Each 25 μL PCR reaction comprised of 2.5 μL DNA (~5 ng/μL), 5 μL each forward and reverse primer (1 pM) and 12.5 μL 2X KAPA HiFi HotStart ReadyMix (Kapa Biosystems, UK). PCR amplification cycles were as follows, initial denaturation at 94 °C for 3 min, followed by 25 cycles of 98 °C for 20 s, X °C for 10 s and 72 °C for 12 s, where X was the annealing temperature optimised at 65, 60 and 66 °C for primer pairs 1, 2 and 3, respectively, and a final extension at 72 °C for 40 s. Amplicons were further processed for library preparation using Illumina’s Nextera XT library preparation kit (Illumina, USA). Sequencing was performed using an Illumina MiSeq desktop sequencer at the Department of Agricultural Biotechnology (a center of excellence in biotechnology, AAU, Anand). Trimming of adaptor sequences was performed using Illumina analysis software V2.5 as recommended by the manufacturer using default parameters.
Sequence data analysis
All 16S rDNA reads were uploaded to the MG-RAST V3.6 open source online server for phylogenetic and functional classification of metagenomics data . Low-quality reads were trimmed using SolexaQA  with default parameters in MG-RAST. Annotations were made against the RDP (Ribosomal Database Project) database with a minimum e value of 1E-5 and minimum identity of 80%. The data were further analysed using Statistical Analysis of Metagenomic Profiles (STAMP v2.1.3) , METAGENassist  and PAST v2.17c . Comparative analysis for taxa in terms of percentage mean relative frequency was performed using STAMP, where Benjamini-Hochberg FDR was used for multiple test corrections to minimise false discovery rates during multiple group comparative analysis. Further, each taxonomic profile from STAMP was uploaded to METAGENassist for analysis where data were filtered for unassigned bacteria and samples were scaled to each other through normalisation using Pareto Scaling (mean-centered and divided by the square root of standard deviation of each variable) . The processed data were subsequently used for correlation analysis. Principal coordinate analysis (PCoA), one-way ANOSIM and clustering were done using PAST. The processed quality filtered sequences were downloaded from the MG-RAST server and used to calculate diversity indices, 2D-PCoA (unweighted UniFrac) and rarefaction curve using QIIME where sequences were clustered at 97% similarity. Moreover, for rarefaction curve sequences were rarefied at 10,000 sequences per sample. Boxplots were generated using BoxPlotR . One-way ANOVA was used to compare three primer pairs while Mann-Whitney test was used to compare differences among two locations within PAST software. To determine the core microbiome of chicken breeds under study, genus abundance > 0.1 and > 1.0% were taken into account and Venn diagrams were generated using Venny 2.1 . To identify chicken line-specific biomarkers at multiple taxonomical levels, the bacterial abundance profile of birds pooled by (i) breed/line and (ii) location were analysed using linear discriminant analysis effect size (LEfSe) [44, 45]. For this analysis, bacterial abundance profiles were calculated at taxonomic levels from phylum to genus in %. For LEfSe analysis, the Kruskal-Wallis test (alpha value of 0.05) and LDA score of > 3.5 were used as thresholds.
Summary statistics of sequences analysed including average OTU numbers detected and microbial diversity covered. For each chicken breed or line, the sequencing reads of all three primer pairs were merged and OTUs were clustered at > 97% similarity using QIIME
Av. sequences/ sample
Av. no. OTUs
Average microbial diversity covered (% Good’s coverage)
Operational taxonomic unit (OTU) occurrence and data rarefaction
Comparison of primer performance
Variation in caecal microbiome structure between chicken breeds and lines
At class level classification, Bacteroidia (p < 8.9E-06) were predominant in all chicken breeds or lines except the Cobb400 reared at location 1, where Clostridia (p < 4.1E-07) were more common (Fig. 3c, Additional file 11). Clostridia were abundant in all chickens raised at location 1. Combined, the Bacteroidia and Clostridia represented the majority of classes within the caecal microbiota. Sequences representative of the class Fusobacteriia (p < 1.4E-07) were common in chickens sampled at location 2, explaining the high occurrence of the phyla Fusobacteria described above.
At genus level, Bacteroides (p < 7.2E-05) was found to be most common in all chicken breeds or lines except the commercial Cobb400 from location 1, where Clostridium (p < 2.1E-03) and unclassified Clostridiales (p < 1.3E-05) presented higher proportions. Alistipes (p < 1.0E-07) were also common in Kadaknath chickens at location 1. The other genera present, and their relative levels are indicated in Fig. 3d (see also Additional file 11).
Numbers of OTUs were high with both of the commercial broiler lines presenting 1411 and 1023 OTUs, followed by Aseel with 1285 and 992 OTUs and Kadaknath with 900 and 665 OTUs (locations 1 and 2, respectively). Comparing both locations, chickens raised at location 1 presented more OTUs than those reared at location 2 (p < 0.0025, Mann-Whitney test; Additional file 12).
The core caecal microbiome, breed-specific biomarkers and correlations among bacteria
We pooled sequencing reads produced using primer pair P2 from each chicken breed or line into a single pool, combining samples from both locations. In total, 35 genera each represented more than 0.1% abundance in the core caecal microbiome of all breeds or lines (Additional files 13 and 14). Twenty one of the 35 genera belonged to the phylum Firmicutes, representing 11 different families. Seven and five genera belonged to the Bacteroidetes and Proteobacteria, respectively with the rest belonging to other diverse phyla. Total of 12 and 10 genera were uniquely detected in Aseel and Cob400 chickens s in excess of this threshold. Comparison of the datasets found Geobacillus, Cyclobacterium, Caldicellulosiruptor, Thermobaculum, Caulobacte, Desulfovibrior and Cytophaga in all Kadaknath birds, but not in any other dataset above the 0.1% threshold, and Slackia, Cronobacter, Phascolarctobacterium, Unclassified Alphaproteobacteria, Oceanimonas, Deferribacter, Tepidimicrobium, Candidatus Phytoplasma, Atopobium, Tannerella, Zunongwangia and Acetobacterium were all found to be specific to Aseel at the same threshold. Two and nine of these genera were identified in both locations for their respective breeds, indicating consistent breed-specific high-level colonisation. Twelve genera were found to exceed the 0.1% threshold in a single commercial broiler line (Cobb400: ten, Ross 308: two) but not an indigenous breed. Just ten genera were detected within the core (conserved) caecal microbiome between all breeds or lines studied when the threshold was increased to 1% abundance (Additional files 13 and 14).
Demand for poultry meat and eggs is increasing dramatically, most notably in South Asia, where a mix of global commercial-type lines and indigenous chicken breeds are kept . As efforts are made to improve the genetic merit of chickens, it is important to identify those genotypes best suited to the varied climatic regions and prevailing production systems, including those which favour a beneficial gut microbiota. The definition of a ‘beneficial’ microbiota is challenging, but comparison of microbiome composition and structure between chicken breeds that have been associated with improved production, health or other welfare traits can provide a valuable source of phenotypic data, with scope to inform genomic analyses to identify host sequence variation associated with microbiome composition. In addition, transplant of ‘beneficial’ microbiota might provide another strategy to improve both production and health in poultry. In inbred mice, transfer of microbiota between mice differing in susceptibility to the enteric pathogen Citrobacter rodentium resulted in a reciprocal transfer of susceptibility and resistance . Here, we selected a single time point late in the broiler production cycle to permit assessment of the outcome of colonisation throughout each chickens’ life. It is important to note that enteric microbial populations will change over time as a consequence of variables such as the development of immune competence and environmental interactions. Future studies will explore fluctuations in microbial diversity as each chicken breed matures.
Here, we employed 16S rRNA amplicon sequencing as a measure of bacterial occurrence within chicken caecal lumen microbiota. We decided to focus on the lumen microbiota in order to standardise sample collection between chickens, and to permit comparison with previous studies by others [e.g. 15]; but recognise that that the mucosal-associated microbiota are likely to present different profiles. 16S rRNA amplicon sequencing has been used widely for classification of diverse microbial communities; however, the bacterial 16S rRNA gene includes nine hypervariable regions and sequences generated using different combinations of these regions commonly present varied profiles of microbial diversity. The optimal choice of hypervariable region(s) and primer combination vary between different ecological populations . Here, we tested three primer sets spanning the V1–V3, V3–V4 or V5–V6 regions of 16S rRNA gene. Rarefaction curves based on the Chao1 index and the number of OTUs detected were comparably close to asymptotic between primer pairs, indicating that the sequences generated per sample were adequate to define and compare the bacterial diversity present within the samples. Further, the mean of Good’s coverage (an alpha diversity index) for all the samples was high (95%). The bacterial diversity, defined as the number of OTUs and species detected, was highest using primer set P2 (V3-V4) and least using P3 (V5-V6). Primer set P2 also illustrated greater species richness using the Chao1 index (accounting for rare OTUs), ACE index (an abundance-based coverage estimator) and PD whole trees (phylogeny-based diversity estimator). Primer set P3 generated a dataset apparently richer in Bacteriodetes sequences with lower total diversity. Based on these observations, we selected primers targeting the V3-V4 region, in agreement with the findings of others where primers targeting the V3, V4, V1–V3, V4–V6 hypervariable regions were compared .
The caecal lumen microbiomes of Indian chicken breeds evaluated here were dominated by sequences representative of the phyla Firmicutes and Bacteroidetes. Combined, these phyla commonly accounted for more than 80% of the total microbial populations detected. Earlier studies have highlighted similar proportions, with Firmicutes commonly dominant [7, 12, 13, 48]. Geographical location exerted a substantial impact on the variation between caecal microbiome populations, including variation in the Firmicutes/Bacteroidetes ratio. In the context of this study, the geographic separation of locations 1 and 2 included husbandry, dietary, climatic and other environmental variables. Such variables have been widely recognised to impact on microbiome structure and diversity . Higher Firmicutes/Bacteroidetes ratios have been associated with human obesity [49, 50], and the reverse has been linked with weight loss . Both Firmicutes and Bacteriodetes have been associated with short chain fatty acid metabolism, although more specifically Firmicutes contribute to butyrate and propionate synthesis, whereas Bacteroidetes primarily synthesise propionate. Bacteriodetes but not Firmicutes produce α-amylase, α-1,2-mannosidase and endo-1,4-β-mannosidase  and are more likely to break down starch and other polymeric substances. Another example of geographic variation was the occurrence of Fusobacteria and Campylobacter, which formed a major component of all chicken breeds raised at location 2 and appeared as a biomarker for the Kadaknath breed at that particular location. Class-level analysis found Clostridia, Gram-positive rod-shaped bacteria including the genera Clostridium, Blautia, Butyrivibrio, Ruminococcus, Roseburia, to be widely abundant. Among these, Blautia, Butyrivibrio and Roseburia have been associated with butyrate production and a positive contribution to the host . The order Clostridiales within the class Clostridia is mainly responsible for short-chain fatty acid metabolism in the chicken caecum . Bacteroidia were also common, including the genera Bacteroides, Alistipes, Parabacteroides, Porphyromonas and others, all of which contribute to propionate production in the caeca . At order level, Clostridiales and Bacteroidales were both abundant (data not shown), although the proportionate representation of Bacteroidales was higher than reported in many previous studies [12, 14, 54], possibly as a consequence of high dietary maize inclusion. Deeper analysis found the genus Bacteroides, gram-negative obligatory anaerobic bacteria of the family Bacteroidaceae, were highly represented in most datasets in line with earlier studies [48, 55]. Bacteroides are generally associated with degradation of polysaccharides, especially starch and glucans [56, 57], and the formation of short-chain fatty acids . The proportion of unclassified bacterial sequences was notably high in many of the samples, encouraging further studies focused on identification and determination of their role(s) or hazards in the enteric microbiota.
While geographic location was a major variable within principle coordinates analysis, separation by chicken breed or line also revealed distinct clusters indicating a host component in microbiome composition, in agreement with previous studies [58, 59]. Similarly, distinct LEfSe biomarkers also defined different chicken breeds or lines. The core caecal lumen microbiome was represented by 48 genera, including Barnesiella, Butyricimonas, Pararevotella, Prevotella, Bacteroides, Clostridium, Ruminococcus, Alistipes, Eubacterium, Bacillus, Lactobacillus, Blautia and Cellulosilyticum, all of which may contribute to chicken food conversion in terms of hydrolysing starch and other macromolecules, and the subsequent formation of short-chain fatty acids via fermentation which are absorbed by the host. All of these genera appeared as breed or line-specific LEfSe biomarkers at one or both locations, indicating a possible host genetic contribution. Potential pathogenic and/or zoonotic organisms within the genera Helicobacter, Campylobacter, Ureaplasma, Eggerthella and Fusobacterium were also detected with abundance > 0.1% in several chicken lines, although clinical disease was not reported . LEfSe analysis further identified some of these genera as breed or line biomarkers, notably associating Kadaknath chickens raised at location 2 with the elevated occurrence of Fusobacterium, Campylobacter, Cronobacter and Enterococcus. The pathogenic bacteria detected could exert both direct and indirect influence on the host and its enteric microbiota. In addition to the risk posed as primary pathogens [61, 62], inflammatory immune responses induced by these pathogens will influence the intestinal environment and its bacterial communities. Further work will be required to determine the relative contributions from the host and these pathogenic bacteria. In contrast, genera associated with beneficial, sometime probiotic effects such as Lactobacillus  were associated as biomarkers with the commercial Cobb400 line. Using 0.1 and 1.0% sequence occurrence cut-off values, a small number of bacterial orders and genera were restricted at higher level occurrence to one or more chicken breeds. Such natural variation may also offer valuable phenotypes amenable to quantitative genetic analysis. Consideration of the indigenous breeds found 11 and 13 genera to represent frequencies in excess of 0.1% for Aseel and Kadaknath, respectively, one and five at a 1.0% cut-off. The occurrence of pathogenic bacteria is of particular relevance. Bacteria of the genus Campylobacter were detected above the 1.0% cut-off in Kadaknath (location 2) and Aseel (locations 1 and 2), but not in Cobb400 or Ross 308. Without undertaking a direct controlled Campylobacter challenge study, it is not possible to comment on the relative resistance or susceptibility of these breeds and lines, although there is some evidence for genetic control of Campylobacter jejuni colonisation in inbred chicken lines . A larger sample size would permit an odds ratio assessment of occurrence and genetic analysis. Culture and detailed genetic characterisation of the Campylobacter species/strains circulating would also be valuable. A controlled study utilising a defined C. jejuni strain and sampling multiple locations within the enteric environment would be required for a definitive assessment of relative resistance or susceptibility.
The relative proportions of several bacteria were correlated in all chickens, regardless of genotype, and represented the core caecal microbiome. Potential pathogens such as those within the genus Fusobacteria were negatively correlated with genera associated with enhanced metabolism such as Butyricicoccus, Pararevotella, Prevotella, Bacteroides, Clostridium, Ruminococcus, Alistipes, Eubacterium, Bacillus, Lactobacillus, Blautia, Cellulosilyticum and Pseudobutyrivibrio, supporting the recognised role for commensal microbiota in chicken health. The characterisation of different bacterial genera as biomarkers associated with specific chicken breeds offers new phenotypes which may be interrogated by quantitative genetic analysis. For example, Lactobacillus and Bacillus showed positive correlation in the Cobb400, but were negatively correlated with most other genera in Kadaknath and presented a mixed response in the Aseel and Ross 308. Interestingly, the Kadaknath populations separated by location presented different LEfSe profiles. While location was clearly a significant variable, it is not possible to confirm whether the two Kadaknath groups represented a single homogenous population or two distinct sub-populations. Future chicken genotyping would be expected to resolve this question. Overall, the Kadaknath chickens consistently presented fewer OTUs and distinct PCoA clusters compared to the Aseel and commercial lines, indicating significant diversity.
The study presented here provides an introduction to chicken breed-specific variation in enteric bacterial occurrence and diversity. The description of variation between global commercial lines and indigenous Indian chicken breeds offers a panel of phenotypes which may be amenable to genetic selection for use in breed improvement. Diversity within breeds may be of greatest interest where there is an opportunity to determine the genetic basis of varied bacterial occurrence. Improved understanding of host-microbiome interactions may support enhanced productivity from low value diets and greater resistance to colonisation by pathogenic and zoonotic organisms.
This research work is an outcome of a collaborative project funded by Department of Biotechnology (DBT, reference BT/IN/Indo-UK/FADH/49/CGJ/2013) India, and the Biotechnology and Biological Sciences Research Council (BBSRC, reference BB/L00478X/1), UK. The funders had no role in study design, data collection and analysis, decision to publish or preparation of the manuscript. This manuscript has been assigned the reference number PPS_01688 by the RVC.
Availability of data and materials
The sequences for this study has been deposited in the NCBI-SRA under the BioProject PRJEB22231 and metagenomes are also available in MG-RAST under accessions 4680540.3, 4680543.3, 4680545.3, 4680547.3, 4680548.3, 4680550.3, 4680552.3, 4680554.3, 4680555.3, 4680557.3, 4680558.3, 4680560.3, 4680562.3, 4680563.3, 4680565.3, 4681303.3-4681332.3, 4692159.3–4692167.3, 4692169.3-4692171.3, 4692175.3-4692185.3, 4692187.3-4692190.3, 4692192.3, 4692195.3, 4692196.3, 4692198.3, 4692200.3-4692203.3, 4692205.3-4692209.3, 4692211.3-4692212.3, 4692214.3, 4692216.3 and 4692369.3.
RJP did data analysis and manuscript writing; ATH and NVP did preliminary data processing, analysis and manuscript correction; TMS and JRT sample processing, DNA isolation and sequencing of samples; PGK, SJJ and TKG did sample collection and manuscript correction. AP, GL, JG and DAH designed experiment and review of manuscript; CGJ, DNR, MR, FMT and DPB involved in acquisition of funding, experimental design and review of the manuscript. All authors read and approved the final manuscript.
This study was carried out using welfare standards consistent with those established under the Animals (Scientific Procedures) Act 1986, an Act of Parliament of the United Kingdom. All protocols were approved by the Ethical Review Panel of Anand Agricultural University (AAU), the Institutional Animal Ethical Committee of the Madras Veterinary College and the Clinical Research Ethical Review Board (CRERB) of the Royal Veterinary College.
Consent for publication
The authors declare that they have no competing interests.
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
- 1.Dixon J, Braun H-J, Kosina P, Crouch JH. Wheat facts and futures 2009. Mexico: Cimmyt; 2009.Google Scholar
- 2.Grace D, Mutua F, Ochungo P, Kruska R, Jones K, Brierley L, Lapar M, Said M, Herrero M, Phuc P: Mapping of poverty and likely zoonoses hotspots. 2012.Google Scholar
- 4.Grace D, Mutua F, Ochungo P, Kruska R, Jones K, Brierley L, Lapar L, Said M, Herrero M, Phuc P. Mapping of poverty and likely zoonoses hotspots. Zoonoses Project 4. Report to the UK Department for International Development. Nairobi: International Livestock Research Institute; 2012.Google Scholar
- 8.Awad WA, Mann E, Dzieciol M, Hess C, Schmitz-Esser S, Wagner M, Hess M. Age-related differences in the luminal and mucosa-associated gut microbiome of broiler chickens and shifts associated with campylobacter jejuni infection. Front Cell Infect Microbiol. 2016;6:154.CrossRefPubMedPubMedCentralGoogle Scholar
- 16.Ahir V, Koringa P, Bhatt V, Ramani U, Tripathi A, Singh K, Dhagat U, Patel J, Patel M, Katudia K. Metagenomic analysis of poultry gut microbes. Indian J Poultry Sci. 2010;45(2):111–4.Google Scholar
- 20.Arora G, Mishra SK, Nautiyal B, Pratap SO, Gupta A, Beura CK, Singh DP. Genetics of hyperpigmentation associated with the Fibromelanosis gene (Fm) and analysis of growth and meat quality traits in crosses of native Indian Kadaknath chickens and non-indigenous breeds. Br Poult Sci. 2011;52(6):675–85.CrossRefPubMedGoogle Scholar
- 29.Klindworth A, Pruesse E, Schweer T, Peplies J, Quast C, Horn M, Glöckner FO. Evaluation of general 16S ribosomal RNA gene PCR primers for classical and next-generation sequencing-based diversity studies. Nucleic Acids Res. 2012. https://doi.org/10.1093/nar/gks808.
- 30.Jaric M, Segal J, Silva-Herzog E, Schneper L, Mathee K, Narasimhan G. Better primer design for metagenomics applications by increasing taxonomic distinguishability. BMC Proc. 2013:1. https://doi.org/10.1186/1753-6561-7-S7-S4.
- 33.Schirmer M, Ijaz UZ, D'Amore R, Hall N, Sloan WT, Quince C. Insight into biases and sequencing errors for amplicon sequencing with the Illumina MiSeq platform. Nucleic Acids Res. 2015; https://doi.org/10.1093/nar/gku1341.
- 35.Chengat Prakashbabu B, Thenmozhi V, Limon G, Kundu K, Kumar S, Garg R, Clark EL, Srinivasa Rao AS, Raj DG, Raman M, et al. Eimeria species occurrence varies between geographic regions and poultry production systems and may influence parasite genetic diversity. Vet Parasitol. 2017;233:62–72.CrossRefPubMedPubMedCentralGoogle Scholar
- 36.Meyer F, Paarmann D, D'Souza M, Olson R, Glass EM, Kubal M, Paczian T, Rodriguez A, Stevens R, Wilke A, et al. The metagenomics RAST server—a public resource for the automatic phylogenetic and functional analysis of metagenomes. BMC Bioinformatics. 2008;9:386.CrossRefPubMedPubMedCentralGoogle Scholar
- 40.Hammer Ř, Harper D, Ryan P. PAST: paleontological statistics software package for education and data analysis–Palaeontol. Electron. 2001;4:9pp.Google Scholar
- 43.Oliveros JC: VENNY. An interactive tool for comparing lists with Venn Diagrams. 2007.Google Scholar
Open Access This article is distributed under the terms of the Creative Commons Attribution 4.0 International License (http://creativecommons.org/licenses/by/4.0/), which permits unrestricted use, distribution, and reproduction in any medium, provided you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons license, and indicate if changes were made. The Creative Commons Public Domain Dedication waiver (http://creativecommons.org/publicdomain/zero/1.0/) applies to the data made available in this article, unless otherwise stated.