Open Access

High-throughput sequencing of 16S rRNA Gene Reveals Substantial Bacterial Diversity on the Municipal Dumpsite

  • Kilaza Samson Mwaikono1, 2Email author,
  • Solomon Maina3,
  • Aswathy Sebastian5,
  • Megan Schilling6,
  • Vivek Kapur2, 6 and
  • Paul Gwakisa2, 4
BMC MicrobiologyBMC series – open, inclusive and trusted201616:145

https://doi.org/10.1186/s12866-016-0758-8

Received: 26 July 2015

Accepted: 2 July 2016

Published: 11 July 2016

Abstract

Background

Multiple types of solid waste in developing countries is disposed of together in dumpsites where there is interaction between humans, animals and the bacteria in the waste. To study the bacteria at the dumpsite and the associated risks, previous studies have focused on culturable, leaving behind a great number of unculturable bacteria. This study focuses on a more comprehensive approach to study bacteria at the dumpsite. Since the site comprised of unsorted wastes, a qualitative survey was first performed to identify the variety of solid waste as this has influence on the microbial composition. Thus, domestic (Dom), biomedical (Biom), river sludge (Riv), and fecal material of pigs scavenging on the dumpsite (FecD) were sampled. Total DNA was extracted from 78 samples and the v4-16S rRNA amplicons was characterized using an Illumina MiSeq platform.

Results

A total of 8,469,294 sequences passed quality control. Catchall analysis predicted a mean of 8243 species per sample. Diversity was high with an average InvSimpson index of 44.21 ± 1.44. A total of 35 phyla were detected and the predominant were Firmicutes (38 %), Proteobacteria (35 %), Bacteroidetes (13 %) and Actinobacteria (3 %). Overall 76,862 OTUs were detected, however, only 20 % were found more than 10 times. The predominant OTUs were Acinetobacter (12.1 %), Clostridium sensu stricto (4.8 %), Proteinclasticum and Lactobacillus both at (3.4 %), Enterococcus (2.9 %) and Escherichia/Shigella (1.7 %). Indicator analysis (P ≤ 0.05, indicator value ≥ 70) shows that Halomonas, Idiomarina, Tisierella and Proteiniclasticum were associated with Biom; Enterococcus, Bifidobacteria, and Clostridium sensu stricto with FecD and Flavobacteria, Lysobacter and Commamonas to Riv. Acinetobacter and Clostridium sensu stricto were found in 62 % and 49 % of all samples, respectively, at the relative abundance of 1 %. None of OTUs was found across all samples.

Conclusions

This study provides a comprehensive report on the abundance and diversity bacteria in municipal dumpsite. The species richness reported here shows the complexity of this man-made ecosystem and calls for further research to assess for a link between human diseases and the dumpsite. This would provide insight into proper disposal of the waste, as well as, limit the risks to human health associated with the dumpsite.

Keywords

Municipal dumpsite Solid waste Molecular diversity 16S rRNA Illumina MiSeq

Background

The amount of solid waste generated has risen due to the increasing urban population in most developing countries [13]. Limited resources and inefficient infrastructure prevent proper waste separation leading to waste remaining in their production sites since it does not get transported to the final dumpsite. The lack of proper waste management systems has also created a dumpsite that includes solid waste, such as plastics, organic waste from households, markets, and abattoirs, agricultural waste, industrial waste, and chemical, pharmaceutical, and biomedical waste. Unattended wastes on dumpsites attract insects, birds, and rodents, as well as, domestic and wild animals. Livestock, such as pigs, goats, and cattle scavenge on this waste. Humans, including children also visit the dumpsite to pick waste without protective gear. The unique, man-made ecosystem in the Arusha municipality in Tanzania, representing conditions in other developing cities, has the potential to cause serious impacts to public health.

Studies on the abundance and diversity of bacteria in environments, especially at interfaces like the dumpsite with human and animal activity, would enrich our understanding of the risks from such interfaces. The presence of disease vectors, such as insects, rodents, and several other small wild animals, at dumpsites can potentially spread pathogens from the dumpsite to other habitats. This highlights the importance of a need for a comprehensive study of the abundance and diversity of bacteria in municipal dumpsites.

Previous studies have only focused on the identification of disease vectors and culturable bacteria on the dumpsite. For example, studies by Ahmed et al., and Onyindo et al., [4, 5] report Periplaneta americanus (Cockroach), Musca domestica (House fly), and Ophyra leucostoma (Black garbage fly) and Stomoxys calcitrans (stable fly) the most prevalent disease vectors on the dumpsite. Awisan et al., [6] found Staphylococcus aureus, Pseudomonas aerugionosa, Klebsiella pneumonia and Escherichia coli as the aerobic and opportunistic bacteria associated to clinical diseases in Irisan dumpsite, and Emmanuel et al., [7] found antibiotic resistant Salmonella spp., Shigella spp., and Vibrio cholerae, Proteus spp., and Pseudomonas spp. from the Utisols dumpsite. However, these methods detect less than 1 % of bacteria found in a particular environment [8, 9]. Other studies have used molecular approaches to identify culturable and unculturable bacteria in a sample, yet, these techniques, such as Sanger sequencing [10, 11], are tedious and inefficient.

In the current study, total genomic DNA was extracted from samples collected from the different types of solid wastes in the dumpsites and the v4 region of the16S rRNA amplicons were sequenced using a high throughput Illumina MiSeq platform. To our knowledge, this is the first report using culture independent approaches and high throughput sequencing to study bacterial abundance and diversity in a municipal dumpsite where interaction between the microbes, animals, disease vectors, and humans is common.

Results

Qualitative survey of the dumpsite

Qualitative survey of the dumpsite revealed the presence of different types of unsorted solid wastes on the dumpsite. Domestic and wild feral animals as well as people were found interacting on the dumpsite. Diversity of animals and solid wastes on the dumpsite is shown in Additional file 1.

Diversities of bacterial communities on dumpsite

A total of 8,469,294 v4 region of 16S rRNA gene sequences of bacteria from 78 solid wastes samples passed all quality control filters. The number of high quality sequences per sample ranged from 329 to 291,482 (mean 108,581, SD 56,553). Catchall analysis of richness predicted an overall mean of 8243 ± 759 species per sample (range 349–21,092, SD 4804). Good’s coverage ranged from 0.9766 to 0.9934 (mean 0.9837, SD 0.0083). Rarefaction curves for different types of solid waste are presented in Additional file 2. The overall diversity of bacteria populations was high with an average inverse Simpson index of 44.21 ± 1.44, Shannon’s evenness average 4.95 ± 0.02, Chao1 richness estimator of 5926 ± 239, and an abundance based coverage estimator (ACE) of 7202 ± 228.

Thirty-five bacterial phyla were detected, however, only seven of these accounted for more than 93 % of all sequences. The predominant phyla were Firmicutes (38 %), Proteobacteria (35 %), Bacteroidetes (13 %), Actinobacteria (3 %), Acidobacteria (2 %), Chloroflexi (2 %) and Spirochaetes (1 %). A large number of phyla were rare and had < 1 % of all sequences (Fig. 1). The unclassified bacteria at phylum level accounted for 2.4 % of all sequences. The heatmap (Additional file 3) shows the abundance and distribution of the predominant bacteria phyla on the dumpsite. When each type of solid waste was separately analysed, Proteobacteria was the most predominant in Biom (37.4 %), Dom (35.7 %) and Riv (42.5 %) solid wastes, while Firmicutes predominated in FecD (59.1 %) only. Majority of the bacterial phyla in all solid waste were rare and contributed < 0.1 % of all sequences. Biom and FecD had only 11 phyla accounted for 98.2 % and 99.7 % of all sequences, respectively. Likewise in Dom and Riv, 12 phyla accounted for 99.6 % and 99.4 % of all sequences, respectively.
Fig. 1

Relative abundance of predominant bacterial phyla on dumpsite. a show the overall predominant taxa in different solid wastes pooled together. Different colours represent types of taxa and their relative abundance. b show levels of abundance of each predominant taxon when each solid waste was analysed separately. Solid biomedical waste (Biom, n =15 domestic solid waste (Dom, n = 33), faecal material of pigs on scavenging on dumpsite (FecD, n = 20) and in river sludge (Riv, n =8). Only phyla in abundance ≥ 0.1 % are shown. Bacterial taxa were assigned at 97 % sequence similarity cut-off level

A total of 76,862 OTUs were detected, yet, only 20 % (15,272/76,864) were identified more than ten times. At family level; Moraxellaceae (12 %), Clostridiaceae-1 (8 %), Ruminococcaceae (5 %), Lachnospraceae (3.7 %) and Lactobacillaceae (3.4 %) were the most abundant. Unclassified bacteria accounted for about 14 % of all sequences at family level. A total of 1350 genera were detected, of which 16 % (220/1350) were unclassified. Acinetobacter was the most abundant genus accounting for (12.1 %) of all sequences followed by Clostridium sensu stricto (4.8 %), Proteiniclasticum (3.4 %), Lactobacillus (3.4 %), Prevotella (2.6 %), Enterococcus (2.9 %) and Escherichia/Shigella (1.7 %) (Fig. 2).
Fig. 2

Abundance of predominant bacteria at genus level in the municipal dumpsite. Bar graph depicts the percentage of sequence reads assigned to each taxon at 97 % sequence similarity cut-off

The core microbiota analyses revealed that none of the OTUs were present across all 78 samples at relative abundance of 1 % or more. Only one OTU, Acinetobacter (assigned sequences 688,875) was found in 62 % (48/78) of samples and Clostridium sensu stricto (assigned sequences 202,494) in 49 % (38/78) of samples. When each type of solid waste was analysed separately, Proteiniclasticum and Acinetobacter were found in 67 % (10/15) of Biom; Clostridium_sensu_stricto in 80 % (16/20) of FecD; Acinetobacter and Proteiniclasticum were found in 73 % (24/33) and 55 % (18/33) of Dom, respectively, and in Riv, 80 % (8/10) of samples had Acinetobacter. Indicator analysis clearly revealed higher affiliation of some bacterial OTUs to specific solid waste on the dumpsite. Thirteen OTUs were significantly associated with Biom (Indicator value ≥ 70 and P ≤ 0.05) amongst which are Halomonas, Alishewanella, and Proteiniclasticum; five were associated to FecD, for example Enterococcus, Bifidobacterium, Clostridium sensu stricto and Cellulosilyticum, and nine were associated with Riv, such as Commamonas, Lysobacter and Flavobacterium (Table 1). None of the OTUs were significantly affiliated with Dom at indicator value ≥ 70 and P ≤ 0.05.
Table 1

Indicator analysis showing bacterial OTUs associated to different solid waste

OTUs

Description (Genus)

#sequences

Indicator value

P-value

 

Biomedical solid waste (Biom)

Otu000140

Idiomarina

6078

89

0.001

Otu000011

Halomonas

83211

86

0.022

Otu000116

Sporolactobacillaceae_incertae_sedis

7683

82

0.023

Otu000035

Tissierella

18780

80

0.007

Otu000080

Alkaliflexus

16213

79

0.004

Otu000136

Saccharofermentans

5945

78

0.005

Otu000124

Alishewanella

13449

78

0.053

Otu000003

Proteiniclasticum

269626

78

0.001

Otu000141

Unclassified Bacteroidetes

7998

75

0.001

Otu000411

Proteiniclasticum

3303

74

0.006

Otu000525

Unclassified Bacteroidetes

346

73

0.001

Otu000567

Pseudomonas

2187

71

0.013

Otu000069

Corynebacterium

18152

70

0.046

 

Faecal material of pigs scavenging on dumpsite (FecD)

Otu000114

Clostridium_sensu_stricto

12444

78

0.039

Otu000477

Unclassified Lachnospraceae

2394

78

0.008

Otu000310

Clostridium_sensu_stricto

2749

77

0.009

Otu000005

Enterococcus

159372

77

0.005

Otu000104

Bifidobacterium

11495

73

0.007

 

River sludge (Riv)

Otu000054

Cloacibacterium

30207

84

0.051

Otu000807

Flavobacterium

2218

84

0.003

Otu000037

Lysobacter

35128

81

0.013

Otu001247

Unclassified Xanthomonadaceae

849

77

0.001

Otu001672

Lacibacter

479

77

0.001

Otu000017

Commamonas

54114

77

0.039

Otu001307

Ferruginibacter

613

75

0.001

Otu000133

Dechloromonas

7673

75

0.039

Otu001398

Novosphingobium

571

74

0.001

Otu000596

Flavobacterium

2057

73

0.002

Otu001167

Niabella

931

72

0.005

None of the OTUs were affiliated to the Dom waste at p ≤ 0.05 and indicator value ≥ 70

Of the 35 bacterial phyla detected, Metastats revealed 11 phyla significantly different between Dom and FecD. Further, at genus level, out of 1428 differentially abundant genera, 173 were significantly different between Dom and FecD. When Biom and FecD were compared, 8 phyla and 144 genera were significantly different. The Biom and Dom comparison revealed no difference in bacteria community at phylum level (P > 0.05), but 9 bacterial family and 39 genera were significantly different. Of the 1428 genera found in Biom and Dom, 16 % (227/1428) were unclassified. The phylum Lentisphaerae, 66 bacterial family and 180 genera were significantly different between Dom and Riv solid waste (Additional file 4).

Comparison of the bacteria community structure and membership between solid wastes

Phylogenetic tree generated using the Yue & Clayton measures as well as the Jaccard index (Additional file 5) were used in the comparison of the bacterial community structure and membership for different solid wastes. Results of the Parsimony test obtained after the phylogenetic analysis of the Yue and Clayton tree ignoring the branch length revealed a significant difference in bacterial community structure between Dom-FecD (P = 0.011) and Dom-Riv (P = 0.028). There was no difference in community structure between Biom-Dom (P = 0.111), Biom-FecD (P = 0.068), Biom-Riv (P = 0.5240 and FecD-Riv (P = 0.100). When branch length was considered, significantly different structures were found between Dom-Riv (P = 0.001) and FecD-Riv (P = 0.034), while none was detected between Biom-Dom, Biom-FecD, Dom-FecD and Biom-Riv (P > 0.05) using Unweighted UniFrac (Table 2). Further, comparison of the community membership using the phylogenetic tree based on the Jaccard index; the parsimony test revealed that Biom-Dom, Dom-FecD and Biom-Riv (P > 0.05) (Table 3) have the same community membership, while in Biom-FecD (P = 0.016), Dom-Riv (P = 0.002) and FecD-Riv (P = 0.002) were different. When Unweighted UniFrac analysis was performed, only Dom-FecD (P = 0.039) had significantly different community membership. Analysis of molecular variance revealed a significant similarity in bacterial community between Biom-Dom (Yue & Clayton, P = 0.475, Jaccard index, P = 0.012), while the rest of the groups were statistically different (P < 0.008) (Tables 3 and 4).
Table 2

Comparison of bacterial community structure between different solid wastes

 

Parsimony

Unweighted UniFrac

Amova

Groups

ParsScore

ParsSig

UWscore

UWSig

FScore

P* - value

Biom-Dom

10

0.111

0.894454

0.052009

2.28212

0.012

Biom-FecD

8

0.068

0.910399

0.085009

3.14737

<0.001

Dom-FecD

10

0.011

0.853535

0.127009

3.38992

0.002

Biom-Riv

7

0.524

0.934552

0.197003

1.92136

0.016

Dom-Riv

6

0.028

0.959495

0.001009

1.80755

0.028

FecD-Riv

6

0.1

0.967847

0.034006

2.46059

0.004

P* = Correction for multiple comparisons (Bonferroni): Significance P- value ≤ 0.0083. Analysis was based on phylogenetic tree generated using Yue & Clayton measure

Table 3

Comparison of bacteria community membership between different solid wastes

 

Parsimony

Unweighted UniFrac

Amova

Groups

ParsScore

ParsSig

UWScore

UWSig

FScore

P* - value

Biom-Dom

11

0.297

0.982725

0.149009

0.988888

0.475

Biom-FecD

7

0.016

0.972863

0.235009

2.24675

< 0.001

Dom-FecD

12

0.08

0.981599

0.039009

2.39594

< 0.001

Biom-Riv

5

0.061

0.98467

0.055007

1.37949

< 0.005

Dom-Riv

5

0.002

0.98451

0.060009

1.57885

0.004

FecD-Riv

4

0.003

0.976693

0.159006

2.25111

0.001

P* = Correction for multiple comparisons (Bonferroni): significance P- value ≤ 0.0083. Analysis was based on phylogenetic tree generated using Jaccard index

Table 4

Summary of good quality sequences and diversity indices of bacteria from different solid waste at species level

bSamples

Valid reads

OTUs

ACEa

Chao1a

Shannona

InvSimpsona

Dom

3,466,427

26,243

7338 ± 230

6788 ± 244

5 ± 1.0

26.9 ± 0.7

Biom

1,706,442

18,994

8218 ± 249

6756 ± 257

5.1 ± 0.01

54.2 ± 1.3

Riv

926,648

15,025

10,422 ± 283

8219 ± 287

5.6 ± 0.01

58.5 ± 1.2

FecD

2,369,595

16,697

6183 ± 214

5073 ± 226

4.5 ± 0.01

31.2 ± 0.9

OTUs Operational taxonomic units, ACE abundance based coverage estimator

aCalculations were performed using Mothur package with an OTU definition at > 97 % sequence similarity

bSamples: Dom—Domestic solid waste; Biom—Solid biomedical waste; Riv—River sludge; FecD—Faecal material of pigs scavenging on dumpsite

Further, Fig. 3a and b are the graphic representation of the PCoA plot based on Bray-Curtis distances. The spatial separation between centers of the clouds of the bacteria community structure of different solid waste using Amova have shown statistical difference between Biom-FecD (P < 0.001), Dom-FecD (P = 0.002) and FecD-Riv (P = 0.004), but the same community structure between Biom-Dom (P = 0.012) which is clearly depicted in a PCoA plot constructed from a pool of bacteria community of the same waste type (Fig. 3b). Bacterial OTUs responsible for the difference in clustering of solid waste types were Halomonas, Acinetobacter and Lactobacillus from Biom and Dom solid waste; Enterococcus and Kurthia in FecD, and Lysobacter in Riv.
Fig. 3

Principal coordinate analysis plots showing clustering of bacterial community from different solid wastes. The PCoA plot was built based on Bray-Curtis dissimilarity distances. a was generated from individual samples from different waste types while in (b) bacteria population from the same waste type was pooled together. Distances between symbols on the ordination plot reflect relative dissimilarities of bacteria community between solid wastes. The OTUs were estimated at 97 % sequence similarity cut-off

Discussion

This study has identified a substantial abundance and diversity of bacteria in a municipal dumpsite in Arusha, Tanzania. The estimated richness (8243 species) per sample and high diversity (InvSimpson index = 44.21) of bacteria on the dumpsite surpasses many of the previous culture based studies. The detection of 76,862 OTUs from 35 bacterial phyla using high throughput sequencing technique gave a more comprehensive look at the bacterial community on the dumpsite. Of the four predominant phyla, Firmicutes was the overall most abundant and includes a variety of gram-positive bacteria. Its particularly dominance in fecal material of pigs is consistent with previous reports by Pajarillo et al., [12] as well as in horse fecal microbiota by Costa et al., [13] and Shepherd et al., [14]. The predominance of Firmicutes in FecD, especially the genus Clostridium sensu stricto may be due to the feeding habits of pigs on fibre from a variety of unsorted solid waste as reported by Middelbos et al., [15] and Yildirim et al.,[16].

The presence of some bacterial genera exclusively to some types of solid waste, justifies a need to sort and treat solid wastes differently. This would deter a possible genetic material exchange, a process which can result in the emergence and re-emergence of new bacteria of public health importance. The revealed affiliation of Enterococcus to FecD is consistent with previous findings in fecal material of pigs [1719]. The potential of Enterococcus in nosocomial infections and multidrug resistance [2022], suggests a need to further examine the antibacterial resistance of isolates from pigs scavenging on dumpsites, and their relation to human and animal pathogens.

The affiliation of Halomonas (1 %) to Biom is consistent with findings in gold mines [23] and in contaminated heavy metals wastes [24], as both environments are rich in chemicals. Halomonas metabolize cyanide [25] and some are reported to carry plasmid pZM3H which confers resistance to different chemicals [24], leading to their application in soil remediation [25, 26]. The fact that these bacteria are predominant, it would be worthwhile to study their roles in the dumpsite, and examining if they contain plasmids that confer resistance to different chemicals. This could lead into their application in the control of chemical pollutants, especially cyanide at the dumpsite.

Proteobacteria was the predominant second overall phylum, and the most abundant in Biom, Dom and Riv solid wastes. Several studies have linked the predominance of this phylum with human and animals diseases [13, 27, 28]. At the genus level, the overall predominance of Acinetobacter (12 %) on the dumpsite is consistent with report by Saini et al., [29] and Hossein et al., [30] in solid waste. Previous studies report some species of Acinetobacter associated with human and animal’s diseases. A case example is the existence of Acinetobacter in clinical isolates from intensive care unit [31], animal with urinary tract infections [32], and as a causative agent of the nosocomial outbreak in Spain [33]. Its predominance in such extreme environment may be linked to their re-counted capacity to detoxify chemicals as well as their multidrug resistance [3335] which would support their survival in an environment with diverse chemicals from unsorted solid wastes.

Escherichia and Shigella spp. were among the predominant genera on the dumpsite. These bacteria were also reported in several culture based studies [3638]. Their capacity to acquire multidrug resistance and their association to human and animal pathogens is well known [3941]. The abundance of these bacteria within the dumpsite with such extraordinary interaction between animals, humans and microbes may be causing potential health risks. The predominance of these bacteria at the dumpsite underscores a need to study on whether a link exists between bacteria on the dumpsite and the known pathogens. Understanding the potential health risks associated with bacteria from the dumpsite would improve the health of not only the humans interacting with the dumpsite, but with all others that may come in contact with them or other animals.

The significant similarity in bacterial community structure and membership between Biom-Dom, Biom—Riv, and FecD—Riv, implies that similar types of bacteria are found in multiple types of solid wastes. This may be attributed to random disposal of unsorted wastes in the same dumpsite and hence, exchange of bacteria between them. Despite all samples were from the same dumpsite whereby exchange of microbes between solid wastes is possible; distinct affiliation of some bacteria OTUs exclusively to some waste was evident. This phenomenon implies that some bacteria have specific nutritional requirement to survive. It would be interesting to study the bacterial community of sorted solid wastes to examine changes in abundance and diversity.

Conclusion

We report an ever rich and diverse bacterial community in Arusha municipal dumpsite. The species richness reported here shows the complexity of this man-made ecosystem and calls for further research to assess for a link between human diseases and the dumpsite. Understanding the role of the bacteria within the dumpsite and bacteria found within different types of waste will provide insight into proper disposal of the waste, as well as, limit the risks to human health associated with the dumpsite.

Methods

Study site and samples

The site for this study was the Arusha municipal dumpsite, where waste from different urban sources is thrown. Sampling was performed in March through June of 2013, whereby prior to sample collection, a qualitative survey was conducted to identify types of most common waste on the dumpsite as these wastes have influence on microbial composition. This comprised waste from households and markets (foods, pampers, clothes, etc.), chemical and biomedical waste (drug containers, used syringes), various plastics and used glassware, waste from abattoirs and brewers, as well as fecal matter from animals scavenging on the dump itself. Samples for this study were the fresh fecal material of pigs scavenging on the dump (FecD, n = 20), domestic solid waste (Dom, n = 33), solid biomedical waste (Biom, n = 15), and run-off water sludge adjoining a nearby river (Riv, n = 8). The core of fresh fecal materials of pigs as well as solid waste and sludge were collected into sterile plastic containers, and, within 1 h, the samples were transported on ice to the laboratory, where total DNA was extracted.

Extraction of total genomic DNA

Total genomic DNA was extracted from about 250 mg of solid waste samples using PowerSoil™ DNA extraction kit (MOBIO Laboratories, Carlsbad, CA) according to the manufacturer’s protocol. Quality and quantity of total DNA was verified with a NanoDrop ND-2000c spectrophotometer (Thermo Scientific) and gel electrophoresis run in 0.8 % agarose and visualized by ultraviolet illumination after staining with gel red™. The DNA was stored at -20oC until further processing.

16S rRNA amplification, Library Construction and Sequencing

The Illumina sequencing preparation guide [42] was used to prepare a pooled amplicon of the v4 region of 16S rRNA gene for sequencing. Primers (515 F/806R) designed for v4 region of 16S rRNA and protocols were adapted from Caporaso [43]. Duplicate reactions were done in PCR master mix reaction in 20 μl AccuPower® Taq PCR PreMix composed of 0.5 μl of 10pmol/μl each for the forward and reverse primers, 17 μl molecular grade water, and 2 μl DNA template. The PCR program was run on GeneAMP™ PCR system 9700 set at 95 oC for 3 min, 35 cycles of 94 oC for 45 s, 50 oC for 60 s and 72 oC for 90 s and a final extension at 72 oC for 10 min. Amplicon quality was visualized using gel electrophoresis, and then pooled and purified using QIAquick® PCR purification kit (Qiagen, German) following manufacturer’s protocol. Purified PCR products were normalized to 120 ng. DNA was quantified using Qubit® dsDNA assay kit in Qubit fluorometer 2.0 (Invitrogen, Life Technologies) and the quality was assessed using Agilent DNA 1000 Chip in Agilent 2100 Bioanalyzer (Agilent Technologies, Waldbronn, Germany.

Library denaturing, dilution, and PhiX control preparation was done as described in the 16S metagenomic sequencing library preparation guide [42]. Libraries were denatured and primers were used according to the method described in Caporaso [43]. Sequencing of the library was performed with the Illumina MiSeq platform (San Diego, USA) using 2 × 250 paired- end chemistry at the BecA—ILRI Hub genomic platform, Nairobi, Kenya.

Sequence data analysis and statistics

The Mothur package algorithms (v1.34.1) were used for both quality control and sequence data analysis [44]. After paired end reads were assembled, sequences were aligned with the Silva 16S rRNA reference database (www.arb-silva.de) [45]. Sequences that were < 239 bp and > 260 bp in length, contained >2 ambiguous base calls or long runs (>8 bp) of homopolymers, or did not align with the correct region were removed. Chimeras were identified using Uchime [46] and eliminated. Taxonomy was assigned using the RDP taxonomy database (http://rdp.cme.msu.edu/index.jsp) [47]. Sequences were binned into operational taxonomic units (OTUs) at 97 % sequence similarity cut-off.

Species richness was assessed with Chao1 [48] and abundance based coverage estimator ACE [49] while evenness and diversity of species were estimated by Shannon [50], jackknife [51] and inverse Simpson [52] indices as well as catchall analysis [53]. All analyses were performed using built-in commands in Mothur v1.34.1. Rarefaction analyses were done at a maximum of 97 % sequence similarity cut-off and was plotted using Phyloseq package [54] in R version 3.1.2. In order to compare bacterial populations between different solid wastes in the same municipal dumpsite, subsampling of sequences from different wastes was done to normalize them for efficient comparison [55]. This consisted of random selection of a number of sequences from each sample consistent to the lowest abundance in all samples. The community membership was compared using the traditional Jaccard index, while population structure was assessed using the Yue & Clayton measure of dissimilarity. Dendrograms were created using Mothur to compare the similarity of bacterial populations among all sample types using both Jaccard index and Yue & Clayton measure which account for the relative abundances in each sample. Figures were generated by FigTree v1.4.2. [56].

To check if the bacterial communities differed significantly between solid wastes, the parsimony [57] and Unweighted UniFrac [58] tests were done. The MOTHUR commands “parsimony” and “Unifrac.Unweighted”, respectively, were applied to the Jaccard and the Yue & Clayton OTU based phylogenetic tree. The statistical significance of the difference in genetic diversity of bacteria community within each solid waste type from the average genetic diversity of both communities pooled together was also assessed using Analysis of Molecular Variance (AMOVA) [59].

The core microbiota analysis was performed in Mothur with command “get.coremicrobiome” and it consisted of identification of OTUs (at relative abundance of 1 %) present in all samples when pooled together and also when each type of solid waste was individually analysed. Indicator analysis [60] was used to test for possible OTUs affiliated to different types of solid waste. Indicator values (IV) ranged from 1 to 100 with higher values for stronger indicators. Though literature considers indicator values > 30 and P-value ≤ 0.05 as good indicators [60], in this study only OTUs with indicator values ≥ 70 and P ≤ 0.05 were judged as having strong affiliation to particular solid waste. The Metastats program [61] was used to identify statistically different OTUs among solid wastes. The shared OTUs file, consensus taxonomy file, and metadata file generated in Mothur v.1.34.1 were imported into METAGENassist [62] where visualization using a heatmap, bar charts and PCA plots was done get more insights on the nature of bacteria present on dumpsite. The BIOM file generated in Mothur was imported into MEGAN5 v5.5.3 [63] where Principal coordinate analysis and relative abundance of different taxa were visualized. A p-value of ≤ 0.05 was considered significant for all comparisons.

Abbreviations

ABCF, African Biosciences Challenge Fund; BecA—ILRI, Biosciences eastern and central Africa—International Livestock Research Institute; MEGAN, MetaGenome ANalyser; OTUs, operational taxonomic units

Declarations

Acknowledgements

We acknowledge the Government of Tanzania through The Nelson Mandela African Institution of Science and Technology for sponsoring this study; Africa Biosciences Challenge Fund (ABCF) for a fellowship to execute laboratory work at Biosciences eastern and Central Africa—International Livestock Research Institute (BecA—ILRI hub) in Nairobi, Kenya. Dr. Rob Skilton for official facilitation while at BecA-ILRI hub, Dr. Francesca Stomeo for her assistance with sequencing, Mr Alan Orth for linking us to ILRI HPC and the Genome Science Centre at Sokoine University of Agriculture for assistance with microbiology and sampling techniques are highly acknowledged.

Funding

This study was funded by the government of Tanzania through the Commission for Sciences and Technology (COSTECH) and the Nelson Mandela African Institute of Science and Technology. Sequencing of DNA and data analyses was partly sponsored by the African Biosciences Challenge Fund (ABCF).

Availability of data and materials

Sequence data discussed in this work have been deposited at the NCBI sequence reads archive (SRA) and assigned with accession numbers. Sequences of bacteria from solid biomedical waste—Biom [GenBank: SRP045926], sequences of bacteria from river sludge near dumpsite—Riv [GenBank: SRP045810], sequences of bacteria from domestic solid waste—Dom [GenBank: SRP046216] and sequences from fecal material of pigs scavenging on dumpsite—FecD [GenBank: SRP046287].

Authors’ contributions

KSM, PSG and VK designed the study; PSG Supervised the study, KSM performed sample collection and DNA extraction; KSM and SM performed sequence library preparation, sequencing, KSM, AS, MS performed data analysis; KSM, SM, AS, MS, VK and PSG were responsible for the draft and final version of the manuscript. All authors read and approved the final draft of the manuscript.

Competing interest

The authors declare no conflict of interest.

Consent for publication

Not applicable

Ethics approval and consent to participate

This study was approved by the research committee of the Nelson Mandela African Institution of Science and Technology. Permits to sample animals and variety of solid wastes on the dumpsite were granted by the Arusha District Veterinary office. Permits to transfer samples between laboratories were given by the Zoosanitary inspectorate services of Tanzania, Arusha (VIC/AR/ZIS/0345) and Veterinary Services under the Ministry of Agriculture Livestock and fisheries of Kenya (RES/POL/VOL.XXIV/506).

Open AccessThis 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.

Authors’ Affiliations

(1)
Department of Science and Laboratory Technology, Dar es Salaam Institute of Technology
(2)
School of Life Sciences and Bioengineering, The Nelson Mandela African Institution of Science and Technology
(3)
BecA-ILRI Hub International Livestock Research Institute
(4)
Genome Sciences Centre, Faculty of Veterinary Medicine, Sokoine University of Agriculture
(5)
Departments of Biochemistry and Molecular Biology, W238A Millennium Science Complex, Penn State University
(6)
Huck Institutes of Life Sciences, Molecular Cellular and Integrative Biosciences, the Pennsylvania State University

References

  1. Mungure MJ. Governance and community participation in Municipal Solid Waste management, case of Arusha and Dar es Salaam Tanzania. M.Sc. In Environmental Management. Aalborg, Denmark: Aalborg University; 2008.Google Scholar
  2. Getahun T, Mengistie E, Haddis A, Wasie F, Alemayehu E, Dadi D, Van Gerven T, Van der Bruggen B. Municipal solid waste generation in growing urban areas in Africa: current practices and relation to socioeconomic factors in Jimma, Ethiopia. Environ Monit Assess. 2012;184(10):6337–45.View ArticlePubMedGoogle Scholar
  3. Guerrero LA, Maas G, Hogland W. Solid waste management challenges for cities in developing countries. Waste Manag. 2013;33(1):220–32.View ArticlePubMedGoogle Scholar
  4. Ahmed A. Insect Vectors of Pathogens in Selected Undisposed Refuse Dumps in Kaduna Town, Northern Nigeria. Sci World J. 2012;6(4):21–26.Google Scholar
  5. Onyido AE, Azubuike J, Amadi ES, Obiukwu MO, Ozumba NA, Ikpeze OO. A survey of public health disease vectors breeding in Refuse Dumps in Onitsha Metropolis, Anambra State Nigeria. N Y J Sci. 2011;4(9):34–9.Google Scholar
  6. Awisan J, Demicais Jr V, Badiang A, Cariño L, Ferrer DM. Aerobic pathogenic and opportunistic bacteria of public health significance in Irisan dumpsite soil. 2013.Google Scholar
  7. Emmanuel I, Joseph N, Kingsley E-I, Egbebor EM, Lawrence E. Antibiotic susceptibility profiles of enteric bacterial isolates from dumpsite utisols and water sources in a rural community in Cross River State, Southern Nigeria. Nat Sci. 2011;9:46–50.Google Scholar
  8. Vandecandelaere I, Matthijs N, Van Nieuwerburgh F, Deforce D, Vosters P, De Bus L, Nelis HJ, Depuydt P, Coenye T. Assessment of microbial diversity in biofilms recovered from endotracheal tubes using culture dependent and independent approaches. PLoS One. 2012;7(6):e38401.View ArticlePubMedPubMed CentralGoogle Scholar
  9. Simon C, Daniel R. Metagenomic analyses: past and future trends. Appl Environ Microbiol. 2010;77(4):1153–61.View ArticlePubMedPubMed CentralGoogle Scholar
  10. Huang LN, Zhu S, Zhou H, Qu LH. Molecular phylogenetic diversity of bacteria associated with the leachate of a closed municipal solid waste landfill. FEMS Microbiol Lett. 2005;242(2):297–303.View ArticlePubMedGoogle Scholar
  11. Chen Y, Dumont MG, Cébron A, Murrell JC. Identification of active methanotrophs in a landfill cover soil through detection of expression of 16S rRNA and functional genes. Environ Microbiol. 2007;9(11):2855–69.View ArticlePubMedGoogle Scholar
  12. Pajarillo EAB, Chae JP, Balolong MP, Kim HB, Seo K-S, Kang D-K. Characterization of the fecal microbial communities of duroc pigs using 16S rRNA gene pyrosequencing. Asian Australas J Anim Sci. 2015;28(4):584–91.View ArticlePubMedPubMed CentralGoogle Scholar
  13. Costa MC, Arroyo LG, Allen-Vercoe E, Stämpfli HR, Kim PT, Sturgeon A, Weese JS. Comparison of the fecal microbiota of healthy horses and horses with colitis by high throughput sequencing of the V3-V5 region of the 16S rRNA gene. PLoS One. 2012;7(7):e41484.View ArticlePubMedPubMed CentralGoogle Scholar
  14. Shepherd ML, Swecker Jr WS, Jensen RV, Ponder MA. Characterization of the fecal bacteria communities of forage-fed horses by pyrosequencing of 16S rRNA V4 gene amplicons. FEMS Microbiol Lett. 2012;326(1):62–8.View ArticlePubMedGoogle Scholar
  15. Middelbos IS, Boler BMV, Qu A, White BA, Swanson KS, Fahey Jr GC. Phylogenetic characterization of fecal microbial communities of dogs fed diets with or without supplemental dietary fiber using 454 pyrosequencing. PLoS One. 2010;5(3):e9768.View ArticlePubMedPubMed CentralGoogle Scholar
  16. Yildirim S, Yeoman CJ, Sipos M, Torralba M, Wilson BA, Goldberg TL, Stumpf RM, Leigh SR, White BA, Nelson KE. Characterization of the fecal microbiome from non-human wild primates reveals species specific microbial communities. PLoS One. 2010;5(11):e13963.View ArticlePubMedPubMed CentralGoogle Scholar
  17. Dang STT, Van Truong D, Madsen H, Dalsgaard A. Survival of faecal indicator bacteria in treated pig manure stored in clay-covered heaps in Vietnam. Vet Microbiol. 2011;152(3–4):374–8.View ArticlePubMedGoogle Scholar
  18. Novais C, Freitas AR, Silveira E, Antunes P, Silva R, Coque TM, Peixe L. Spread of multidrug-resistant Enterococcus to animals and humans: an underestimated role for the pig farm environment. J Antimicrob Chemother. 2013;68(12):2746–54. dkt289.View ArticlePubMedGoogle Scholar
  19. Freitas AR, Coque TM, Novais C, Hammerum AM, Lester CH, Zervos MJ, Donabedian S, Jensen LB, Francia MV, Baquero F. Human and swine hosts share vancomycin-resistant Enterococcus faecium CC17 and CC5 and Enterococcus faecalis CC2 clonal clusters harboring Tn1546 on indistinguishable plasmids. J Clin Microbiol. 2011;49(3):925–31.View ArticlePubMedPubMed CentralGoogle Scholar
  20. Arias CA, Murray BE. The rise of the Enterococcus: beyond vancomycin resistance. Nat Rev Microbiol. 2012;10(4):266–78.View ArticlePubMedPubMed CentralGoogle Scholar
  21. Lebreton F, van Schaik W, McGuire AM, Godfrey P, Griggs A, Mazumdar V, Corander J, Cheng L, Saif S, Young S: Emergence of epidemic multidrug-resistant Enterococcus faecium from animal and commensal strains. mBio. 2013;4(4):e00534-00513.Google Scholar
  22. Marcadé G, Micol J-B, Jacquier H, Raskine L, Donay J-L, Nicolas-Viaud S, Rouveau M, Ribaud P, Dombret H, Leclercq R. Outbreak in a haematology unit involving an unusual strain of glycopeptide-resistant Enterococcus faecium carrying both vanA and vanB genes. J Antimicrob Chemother. 2013;376:1–6.Google Scholar
  23. Lin Y, Fan H, Hao X, Johnstone L, Hu Y, Wei G, Alwathnani HA, Wang G, Rensing C. Draft genome sequence of Halomonas sp. strain HAL1, a moderately halophilic arsenite-oxidizing bacterium isolated from gold-mine soil. J Bacteriol. 2012;194(1):199–200.View ArticlePubMedPubMed CentralGoogle Scholar
  24. Dziewit L, Pyzik A, Matlakowska R, Baj J, Szuplewska M, Bartosik D. Characterization of Halomonas sp. ZM3 isolated from the Zelazny Most post-flotation waste reservoir, with a special focus on its mobile DNA. BMC Microbiol. 2013;13(1):59.View ArticlePubMedPubMed CentralGoogle Scholar
  25. Khamar Z, Makhdoumi A, Mahmudy Gharaie MH. Remediation of cyanide from the gold mine tailing pond by a novel bacterial co-culture. Int Biodeterioration Biodegradation. 2015;99:123–128.View ArticleGoogle Scholar
  26. Achal V, Pan X, Zhang D. Bioremediation of strontium (Sr) contaminated aquifer quartz sand based on carbonate precipitation induced by Sr resistant Halomonas sp. Chemosphere. 2012;89(6):764–8.View ArticlePubMedGoogle Scholar
  27. Suchodolski JS, Markel ME, Garcia-Mazcorro JF, Unterer S, Heilmann RM, Dowd SE, Kachroo P, Ivanov I, Minamoto Y, Dillman EM. The fecal microbiome in dogs with acute diarrhea and idiopathic inflammatory bowel disease. PLoS One. 2012;7(12):e51907.View ArticlePubMedPubMed CentralGoogle Scholar
  28. Suchodolski JS, Xenoulis PG, Paddock CG, Steiner JM, Jergens AE. Molecular analysis of the bacterial microbiota in duodenal biopsies from dogs with idiopathic inflammatory bowel disease. Vet Microbiol. 2010;142(3):394–400.View ArticlePubMedGoogle Scholar
  29. Saini S, Das BK, Kapil A, Nagarajan SS, Sarma R. The study of bacterial flora of different types in hospital waste: evaluation of waste treatment at AIIMS Hospital, New Delhi. Southeast Asian J Trop Med Public Health. 2004;35(4):986.PubMedGoogle Scholar
  30. Hossain MS, Rahman NNNA, Balakrishnan V, Puvanesuaran VR, Sarker MZI, Kadir MOA. Infectious risk assessment of unsafe handling practices and management of clinical solid waste. Int J Environ Res Public Health. 2013;10(2):556–67.View ArticlePubMedPubMed CentralGoogle Scholar
  31. Omer MI, Gumaa SA, Hassan AA, Idris KH, Ali OA, Osman MM, Saleh MS, Mohamed NA, Khaled MM. Prevalence and resistance profile of acinetobacter baumannii clinical isolates from a private hospital in Khartoum, Sudan. Am J Microbiol Res. 2015;3(2):76–9.Google Scholar
  32. Pomba C, Endimiani A, Rossano A, Saial D, Couto N, Perreten V. First report of OXA-23-mediated carbapenem resistance in sequence type 2 multidrug-resistant acinetobacter baumannii associated with urinary tract infection in a cat. Antimicrob Agents Chemother. 2014;58(2):1267–8.View ArticlePubMedPubMed CentralGoogle Scholar
  33. Mosqueda N, Espinal P, Cosgaya C, Viota S, Plasensia V, Álvarez-Lerma F, Montero M, Gómez J, Horcajada JP, Vila J. Globally expanding carbapenemase finally appears in Spain: nosocomial outbreak of Acinetobacter baumannii producing plasmid-encoded OXA-23 in Barcelona, Spain. Antimicrob Agents Chemother. 2013;57(10):5155–7.View ArticlePubMedPubMed CentralGoogle Scholar
  34. Franzetti A, Gandolfi I, Fracchia L, Van Hamme J, Gkorezis P, Marchant R, Banat IM. Biosurfactant use in heavy metal removal from industrial effluents and contaminated sites. Biosurfactants. 2014;159:361.Google Scholar
  35. Zarrilli R, Pournaras S, Giannouli M, Tsakris A. Global evolution of multidrug-resistant Acinetobacter baumannii clonal lineages. Int J Antimicrob Agents. 2013;41(1):11–9.View ArticlePubMedGoogle Scholar
  36. Rastogi V, Rastogi P, Bhatia S. Bacteriological profile of biomedical waste: management guidelines. J Indian Acad Forensic Med. 2011;33(2):145–8.Google Scholar
  37. Anitha J, JAYRAAJ IA. Isolation and identification of bacteria in biomedical wastes (BMW). Int J Pharm Pharmaceutical Sci. 2012;4(5):386–8.Google Scholar
  38. Oyeleke S, Istifanus N: The microbiological effects of hospital wastes on the environment. Afr J Biotechnol. 2009;8(7):1253–1257.Google Scholar
  39. Doumith M, Dhanji H, Ellington MJ, Hawkey P, Woodford N. Characterization of plasmids encoding extended-spectrum β-lactamases and their addiction systems circulating among Escherichia coli clinical isolates in the UK. J Antimicrob Chemother. 2012;67(4):878–85.View ArticlePubMedGoogle Scholar
  40. García A, Fox JG, Besser TE. Zoonotic enterohemorrhagic Escherichia coli: a One Health perspective. Ilar J. 2010;51(3):221–32.View ArticlePubMedGoogle Scholar
  41. Fritah S, Lhocine N, Golebiowski F, Mounier J, Andrieux A, Jouvion G, Hay RT, Sansonetti P, Dejean A. Sumoylation controls host anti‐bacterial response to the gut invasive pathogen Shigella flexneri. EMBO Rep. 2014;15(9):965–72.View ArticlePubMedPubMed CentralGoogle Scholar
  42. Illumina: 16S Metagenomic Sequencing Library preparation Guide: (http://support.illumina.com/downloads/16s_metagenomic_sequencing_library_preparation.html). 2013. Accessed 6 Mar 2014.
  43. Caporaso JG, Lauber CL, Walters WA, Berg-Lyons D, Lozupone CA, Turnbaugh PJ, Fierer N, Knight R. Global patterns of 16S rRNA diversity at a depth of millions of sequences per sample. Proc Natl Acad Sci. 2011;108(Supplement 1):4516–22.View ArticlePubMedGoogle Scholar
  44. Schloss PD, Westcott SL, Ryabin T, Hall JR, Hartmann M, Hollister EB, Lesniewski RA, Oakley BB, Parks DH, Robinson CJ. Introducing mothur: open-source, platform-independent, community-supported software for describing and comparing microbial communities. Appl Environ Microbiol. 2009;75(23):7537–41.View ArticlePubMedPubMed CentralGoogle Scholar
  45. Quast C, Pruesse E, Yilmaz P, Gerken J, Schweer T, Yarza P, Peplies J, Glöckner FO. The SILVA ribosomal RNA gene database project: improved data processing and web-based tools. Nucleic Acids Res. 2013;41(D1):D590–6.View ArticlePubMedGoogle Scholar
  46. Edgar RC, Haas BJ, Clemente JC, Quince C, Knight R. UCHIME improves sensitivity and speed of chimera detection. Bioinformatics. 2011;27(16):2194–200.View ArticlePubMedPubMed CentralGoogle Scholar
  47. Cole JR, Wang Q, Fish JA, Chai B, McGarrell DM, Sun Y, Brown CT, Porras-Alfaro A, Kuske CR, Tiedje JM. Ribosomal Database Project: data and tools for high throughput rRNA analysis. Nucleic Acids Res. 2014;42(D1):D633–42.View ArticlePubMedGoogle Scholar
  48. Chao A. Nonparametric estimation of the number of classes in a community. Scand J Stat. 1984;11:265–70.Google Scholar
  49. Chao A, Lee S-M. Estimating the number of classes via sample coverage. J Am Stat Assoc. 1992;87(417):210–7.View ArticleGoogle Scholar
  50. Chao A, Shen T-J. Nonparametric estimation of Shannon’s index of diversity when there are unseen species in sample. Environ Ecol Stat. 2003;10(4):429–43.View ArticleGoogle Scholar
  51. Heltshe JF, Forrester NE. Estimating species richness using the jackknife procedure. Biometrics. 1983;39(1):1–11.View ArticlePubMedGoogle Scholar
  52. Hunter PR, Gaston MA. Numerical index of the discriminatory ability of typing systems: an application of Simpson’s index of diversity. J Clin Microbiol. 1988;26(11):2465–6.PubMedPubMed CentralGoogle Scholar
  53. Bunge J. Estimating the number of species with CatchAll. In: Pacific Symposium on Biocomputing: 2011. World Sci. 2011;PBS 2011:121–130.Google Scholar
  54. McMurdie PJ, Holmes S. phyloseq: an R package for reproducible interactive analysis and graphics of microbiome census data. PLoS One. 2013;8(4):e61217.View ArticlePubMedPubMed CentralGoogle Scholar
  55. Gihring TM, Green SJ, Schadt CW. Massively parallel rRNA gene sequencing exacerbates the potential for biased community diversity comparisons due to variable library sizes. Environ Microbiol. 2012;14(2):285–90.View ArticlePubMedGoogle Scholar
  56. Rambaut A: FigTree v1. 3.1: Tree figure drawing tool. Website: http://tree bio ed ac uk/software/figtree. 2009. [Accessed 10 Jul 2015].
  57. Schloss PD, Handelsman J. Introducing TreeClimber, a test to compare microbial community structures. Appl Environ Microbiol. 2006;72(4):2379–84.View ArticlePubMedPubMed CentralGoogle Scholar
  58. Lozupone CA, Hamady M, Kelley ST, Knight R. Quantitative and qualitative β diversity measures lead to different insights into factors that structure microbial communities. Appl Environ Microbiol. 2007;73(5):1576–85.View ArticlePubMedPubMed CentralGoogle Scholar
  59. Michalakis Y, Excoffier L. A generic estimation of population subdivision using distances between alleles with special reference for microsatellite loci. Genetics. 1996;142(3):1061–4.PubMedPubMed CentralGoogle Scholar
  60. Dufrêne M, Legendre P. Species assemblages and indicator species: the need for a flexible asymmetrical approach. Ecol Monogr. 1997;67(3):345–66.Google Scholar
  61. White JR, Nagarajan N, Pop M. Statistical methods for detecting differentially abundant features in clinical metagenomic samples. PLoS Comput Biol. 2009;5(4):e1000352.View ArticlePubMedPubMed CentralGoogle Scholar
  62. Arndt D, Xia J, Liu Y, Zhou Y, Guo AC, Cruz JA, Sinelnikov I, Budwill K, Nesbø CL, Wishart DS. METAGENassist: A comprehensive web server for comparative metagenomics. Nucleic Acids Res. 2012;40(Web Server issue):W88–95. gks497.View ArticlePubMedPubMed CentralGoogle Scholar
  63. Mitra S, Stärk M, Huson DH. Analysis of 16S rRNA environmental sequences using MEGAN. BMC Genomics. 2011;12 Suppl 3:S17.View ArticlePubMedPubMed CentralGoogle Scholar

Copyright

© The Author(s). 2016