A Review of Basic Bioinformatic Techniques for Microbial Community Analysis in an Anaerobic Digester

Abstract

Biogas production involves various types of intricate microbial populations in an anaerobic digester (AD). To understand the anaerobic digestion system better, a broad-based study must be conducted on the microbial population. Deep understanding of the complete metagenomics including microbial structure, functional gene form, similarity/differences, and relationships between metabolic pathways and product formation, could aid in optimization and enhancement of AD processes. With advancements in technologies for metagenomic sequencing, for example, next generation sequencing and high-throughput sequencing, have revolutionized the study of microbial dynamics in anaerobic digestion. This review includes a brief introduction to the basic process of metagenomics research and includes a detailed summary of the various bioinformatics approaches, viz., total investigation of data obtained from microbial communities using bioinformatics methods to expose metagenomics characterization. This includes (1) methods of DNA isolation and sequencing, (2) investigation of anaerobic microbial communities using bioinformatics techniques, (3) application of the analysis of anaerobic microbial community and biogas production, and (4) restriction and prediction of bioinformatics analysis on microbial metagenomics. The review has been concluded, giving a summarized insight into bioinformatic tools and also promoting the future prospects of integrating humungous data with artificial intelligence and neural network software.

1. Introduction

With the increase in energy demand and rise in fuel prices in recent years, renewable sources of energy have caught much attention lately. Environmental pollution caused by burning fossil fuels is reduced by extensive use of renewable energy sources. Environment protection and reducing the energy crisis simultaneously could be attained by using and upgrading various feedstocks such as food waste, agricultural residues, solid waste, algae, etc. Feedstock-based anaerobic digesters produce biogas and are an efficient way to overcome the problems related to fossil fuel burning. Biogas containing diverse microbial communities in digesters is critical in maintaining stable activities and the efficient development of methane [1,2,3,4,5]. Figure 1 shows the process and microbial community involved in anaerobic digestion. Biogas engendered from anaerobically digested biological wastes by metabolism of wide-ranging anaerobically active microbes, especially bacteria and methanogenic archaea, is one of the striking energy carriers that is sustainable and renewable and for which substantial research is the need of the hour. The monitoring methods of standard operational parameters, such as temperature, pH, gas composition, alkalinity and volatile fatty acid (VFA) concentration, has enhanced the bio-methanation process and expanded AD systems at the industrial scale [6,7,8,9]. These parameters reveal the present condition of the process, but the exact dynamics, composition, metabolism, or activity of the anaerobic system is still a mystery to be revealed. [10,11].

Anaerobic digestion is divided into four phases, of which the first three phases, i.e., hydrolysis, acidogenesis, and acetogenesis, are carried out by bacterial populations and the last phase, i.e., methanogenesis, occurs by a specific group of archaea. An AD process is a complex metabolic system where different types of microbial flora interact with each other to give a different kind of end-product at different levels. This kind of interaction is called syntropic interaction with specific anaerobic functionality and has been explored using several molecular biology tools [12]. The functional group genes were targeted by designing a primer for the 16S rRNA gene. Using these primers as a phylogenetic marker, distinct functional group genes were targeted, and analyses were conducted on that basis. By means of molecular techniques it was evident that microbial flora in anaerobic digesters is unexplored, so deeper studies must be carried out using bioinformatics and molecular tools [13]. The efficiency of gas production depends on the steady operation of the anaerobic digester henceforth maintained by the intricate microbial population [1,2]. Methane production can be increased through the strategic enrichment of hydrolytic bacteria, fermentation bacteria, and advantageous methanogens through functional separation at various time points [14]. It is possible to inhibit the development of harmful metabolites and promote the production of chemicals essential for the continuity of anaerobic cell membranes with the help of micro-oxygen, which improves the overall performance of AD [15]. With respect to taxonomic arrangements, interaction systems, metabolic links, species similarity, and diversity, the microbial consortia must be evaluated so that the anaerobic digestion of the organic waste can produce a high amount of biogas. In the last ten years, the advancement of high-performance sequencing technologies and cost-reduction has rendered it possible to use bioinformatic techniques in aerobic fermentation tanks to analyze metagenomic data in microbial communities. Many techniques from the field of bioinformatics, which are based on statistics, data mining, and artificial intelligence, such as Genovo, MG-RAST, MetaVelvet, etc, [16,17,18], have been used to enhance knowledge of microbial populations in anaerobic digesters working under different operational systems. Luo et al. reported that anaerobic microbial populations in biogas digesters show a strong link between functional gene form and taxonomic pattern [19]. Zhang J et al. revealed that with the help of functional genes, analysed by network-based techniques and metagenomic studies, many metabolic pathways can be estimated. Similar findings were reported by Ye L et al., who showed the overall microbial metabolic pathways of anaerobes in biogas producing digesters [20,21]. In addition, it has been found that operational parameters such as organic loading rate, type of feedstock, temperature, design of digester, alkalinity, concentration of free ammonia, and hydraulic retention time; influence both functional genes and taxonomic patterns [22,23,24]. Efficient analysis of the fundamental biological systems of anaerobic populations, viz., bacteria and archaea, in anaerobic bioreactors through wide-ranging sequenced data and also using these data to enhance the production of biomethane, is still a herculean task. Unavailability of accurate gene-sequencing technology, presence of very complex microbial populations in anaerobic tanks, operational errors during analysis of microbial dynamics, disintegrated analysis of microbial population data, and the high cost of metagenomic analysis, make the overall study of microbial communities a major challenge [25,26]. Metagenomic study of microbial populations is not only conducted for anaerobic digesters but also it has been conducted for numerous other habitats such as contaminated sites, sea, soil hydrocarbon contaminated sites, and biohydrogen reactors [27,28,29,30]. Many researchers have been investigating microbial populations in anaerobic digesters at different operational parameters. It has been found that although much research is being conducted in the field of metagenomics, the data generated are not consolidated and are very diverse and scattered. Therefore, a suitable review in this domain will help to build a consolidated knowledge. This review has been designed to provide overall information of the bioinformatic tools and approaches used for metagenomic studies of microbial communities in anaerobic digesters. The paper has been divided into various sections. Section 2 summarises the metagenomic studies conducted so far in anaerobic digesters with different kinds of feedstock. This section further presents methods of DNA isolation and DNA sequencing. Figure 2 shows the basic flow chart of the metagenomic analysis process. Zhang et al. [20] reported that metagenomic studies are conducted in terms of operational parameters, country of origin, and type of feedstock on which the anaerobic digester is working. Countries, for example, Singapore, UK, USA, Korea, China, Germany, Denmark, etc., have been working towards metagenomic studies of anaerobic digesters. Most of the microbial community analysis has been conducted on food waste as feedstock, followed by sludge, manure, and horticultural and agricultural residue. Moreover, mesophilic digesters are more explored than thermophilic digesters for microbial population analysis [31]. With high temperature, thermophilic AD is more advantageous than mesophilic as frothing in the digester is reduced and also due to high temperature, degradability of complex polymers is enhanced. Due to this, enhanced biomethane is obtained and organic loading rate is also increased [32,33,34].

2. Methods of DNA Isolation and DNA Sequencing

Isolation of DNA is the first and most important step in studying the microbial community. DNA isolation is complex process and to isolate purified DNA is a major task as it depends on many factors such as sampling method. Therefore, selection of the correct sampling method is very important and hence the designing of an experiment is conducted on the basis of sampling for DNA isolation. Repeatability of the experiment is needed to get more precise results and experimental errors are also reduced. The sampling method should be kept constant throughout the experiment to reduce errors [35]. Digestate is mostly collected at a fixed interval of days so that a complete consortium of microorganism can be analysed. The sample should be pre-treated to remove impurities that could create errors in the DNA isolation and PCR processes. Pre-treatment of the digestate should be carried out by rinsing it with phosphate buffer [36]. A general method for isolation of DNA is followed by centrifuging the digestate and collecting the supernatants. The step-by-step method of precipitation of DNA using phenol chloroform is followed. Although most of the DNA-isolation process is carried out using commercially available DNA kits, Bin dong et al. reported that the most efficient method of DNA isolation is the enzymatic disruption method.

Table 1. Commonly used DNA-isolation kits.

Brand Name	Country Name	DNA Isolation Kit	Usage Percentage	Reference
Clontech	USA	PCR reaction mix	2%	[40]
Zymo Research	USA	ZR soil microbe DNA kit	2%	[44]
Felix bio-tech	USA	DNA extraction kit	2%	[45]
Intron biotechnology	Korea	I-genomic BYF DNA extraction kit	2%	[46]
Magtration System 6GC, Precision System Science	Japan	Automated nucleic acid kit	6%	[38,47,48]
Macherey-Nagel	Germany	NucleoSpin Tissue kit+NucleoSpin soil kit	6%	[49]
OMEGA	USA	E.Z.N.A Soil DNA kit	6%	[10,50]
Q-Bio gene	Australia, Carlsbad, CA, USA	Fast DNA SPIN kit for soil	25%	[51]
MP Biomedicals	Illkirch, France, Australia, Germany, USA			[52]
MoBio Laboratories	USA	MoBio PowerSoil DNA extraction kit	43%	[53,54]
-	-	CTAB (cetyltrimethylammonium bromide) based DNA extraction method	6%	[37,55,56,57,58,59]

lists the DNA extraction processes utilized by anaerobic digesters in microbial cultures. OMEGA, Precision System Science, MP Biomedicals, Intron Technology, and MOBio Laboratories are some of the leading companies that provide commercial DNA-isolation kits. After DNA isolation is completed, the quality of the DNA is checked through Agarose Gel electrophoresis. The quantity and purity of the DNA could be checked by measuring the absorbance at 260 and 280 nm, respectively, using both an UV spectrophotometer and NanoDrop [37]. Purified DNA is stored at −20 °C in Tris-EDTA for further use. Typically, Pure DNA shows approx. 1.9 absorbance at 260 nm and 50 ng/nL concentration [38]. Microbial community analysis of anaerobic digesters on the basis of quality and quantity is conducted rapidly with the help of next-generation sequencing methods. Sun et al. reported that pyrosequencing programs such as Roche GS FLX 454 is the most frequently used platform for microbial community analysis of anaerobic digesters [39]. Some of the frequently used platforms, such as ABI SOLiD™ and reagents for ABI analysis, are platforms for short-read DNA sequencing, upon which next-generation sequencing depends. Few NGS methods are based on platforms such as the Illumina (Solexa) sequencing platform and sequencing kit (Ion PGM™ Hi-Q™) coupled with a sequencer (Ion PGM™), which is controlled by software, namely Torrent Suite™ [40]. A new generation of sensing technology uses nanopores—nano-scale holes—embedded in high-tech electronics, to perform comprehensive molecular analyses. Sequencing at the sample source using portable MinION and Flongle devices use inbuilt EPI2ME data-analysis workflows for real-time species ID and AMR profiling. However, it is used much less frequently than Illumina to analyse the microbial community of AD and is most often used as an adjunct to Illumina [41,42,43]. Although there are various methods of DNA sequencing, there are various errors that must be addressed. For example, by removing the error from the de-novo genome assembly, convolutions of the downstream estimations will be reduced. Many error-correction tools are under development and are being used to reduce the complexity of the metagenomic analysis process [12].

3. Analysis of Anaerobic Microbial Communities Using Bioinformatic Tools

Metagenomic data are analysed through various bioinformatic tools, which are different for different steps involved in the whole procedure of metagenomic data analysis. The overall analysis of microbial populations involves different facets of analysis. First of all, raw sequences are pre-treated using several bioinformatics softwares. The pre-treated sequences are now sequentially analysed on different levels. As shown in Figure 3, the pre-treated sequences undergo different analyses. Operational taxonomic unit-clustering analysis, taxonomical compositional analysis, analysis of alpha diversity, analysis of similarity and difference in the microbial community, functional gene analysis, and statistical analysis are different attributes which require conforming techniques and platforms. Chowdhary and Kumar [29] described various software for metagenome data analysis with attached web functions and links. Similarly, Ju and Zhang [35] focused on frameworks and computational resources available for metagenomic bioinformatics analysis, covering several data-processing functions, including pre-treatment, binning, annotation, and assembly. Nevertheless, these reported computer software and platforms [21,44] have not been fully applied to anaerobic microbial metagenomics in biogas-producing digesters as the samples were extracted from various microbial habitats such as soil, ocean, sewage, biofilms, and bioreactors, etc.

3.1. Pre-Processing of Raw Sequences

Upon the collection of metagenome data (direct NGS reads), fresh sequence pre-treatment is quite a crucial move for achieving high-quality readings for the downstream analysis. A sequenced pre-treatment usually includes (i) removal of linkers and adapters, (ii) chimera’s exclusion and its replication, and (iii) demultiplexing of barcoded samples and quality control. There are various software tools (

Table 2. Different analysis type and the common software used.

Analysis Type	Name of Software	Reference
Measurement and analysis of performance, systemic metabolic processes, annotation of genomes, study of principal coordinates, statistics based on canonical correspondence, data from filtered pyrosequencing runs.	MG-RAST	[65]
Removing chimaera sequences, ensuring high quality, the assessment of variety, richness, and abundance of base coverage, study of principal coordinates, analysing Good’s scope of coverage, Alignment and quality control of sequences, sample size normalisation	MOTHUR	[52]
Shortening of reads	Trimmomatic software	[62]
Shortening and aligning reads	HMMER	[67]
Identified sequence reads	ARB rRNA database	[67]
Assembling of genome	CLC Genomics workbench	[57,62,63]
Combination of end-pair reads	FLASH	[17,68]
Analysis of microbial population relation	MetaMIS	[17,65,68]
Interaction network topology analysis	Gephi	[17]
CLUSTAL_X	Iterative sequence alignments; gap editing	[17]
PAST	Illustration of beta-diversity metrics	[57,62]
mPUMA and Trinity	Assembly and processing sequences	[65]
Chimera Slayer	Chromosome complexes elimination.	[69]
ClustalW	Alignment of sequences	[49]
INFERNAL aligner	Alignment of various clean sequences	[10,49,70]
SAMS	Evaluation of the quality of sequences	[70]
GenDB genome annotation system	Long-read assembly and functional annotation	[10]
Regano	Code-sequence prediction	[60,71]
Pipeline	Aligning, trimming, and sorting sequences; analysing biodiversity; naming sequences by taxonomy	[60]
RDP (Ribosomal Database Project)	Differences in community architecture	[69]
Fast UniFrac	Illustrate co-relations between microbial structure and attributes	[69]
CANOCO	Identify causes and effects of microbial communities on reactor efficiency	[69]
XLSTAT	Comparison of taxonomies between two samples using pairwise statistics	[72]
STAMP	MG-RAST	[65]

) which have been used in last ten years for this purpose. The software tools and bioinformatics platforms include UCHIME [10,50,60], RDP tools [61], ACE Pyrotag Pipeline (APP) [52], HMMER [62], MG-RAST [63], Chimera Slayer [64], and Trimmomatic software [65]. MOTHUR and QIIME (http:/qiime.org/, accessed on 13 November 2022) are the two most-used platforms for deactivating metagenome data, while UCHIME is the most popular tool for eliminating and verifying chimaeras from raw sequences. The sequencing reads are organised by barcode and then all primers and barcodes are cleared after primary processing is complete. The barcoding is generally carried out using the Nextera XT Index Kit v.2 (Illumina, San Diego, CA, USA) [66]. Quality control must be carried out for further analysis. For example, all sequences under 150 bp are disregarded by the RDP Pyrosequencing Pipeline’s quality control methodology, and samples containing more than one unknown nucleotide, forward primer mismatches, or have poor base quality ratings (Phred quality scores less than 25) require additional investigation [61]. In addition, the pre-cluster method is typically used to combine sequences with 1 bp difference.

3.2. Analysis of OTU Clustering

Various clean sequences are aligned with sequence aligners such as Py-NAST [55], INFERNAL aligner [61], MOTHUR [71], ClustalW [36], and MUSCLE [67] in conjunction with the bacterial and archaeal database SILVA [60]. Figure 4 is a model diagram for functional hits of COG. Usearch software is then used to cluster linked sequences into operating taxonomy units (OTUs) using the average adjacent clustering algorithm (Usearch-global command) [72]. Ninety-seven percent sequence similarity is typical for sequence classifiers such as RDP Bayesian classifier [61], UCLUST-RDP classifier [73], MEGA/MEGA5 MEGA 10, and MEGA X [36,60,74]. Of note, the normalization of sample size can be accomplished using a program such as MOTHUR by re-sampling the same number of readings for all samples depending on the smallest sample size. OTU-based analysis of bacterial and archaeal sequences can then be further performed. For example, utilizing the R-Venn-Diagram programme, Yun et al. [75] examined the microbial community dynamics of ammonia inhibition during its mitigation by internal dilution in high-rate anaerobic digestion of food-waste leachate. Begmatov et al. studied OTUs for taxonomic analysis by searching against the SILVA v.138 rRNA sequence database via VSEARCH v. 2.14.1 algorithm 28. Similarly, MiDAS 414, a recently developed reference database of full length 16S rRNA genes for waste-water treatment plants was used for species-level taxonomical analysis [66].

3.3. Analysis of Alpha Diversity

A 97% cut-off level of OTUs was used for diversity-indices calculation using Usearch v.1127 [66]. As a rule, the Chao index characterizes the richness of a community, while the Shannon and Simpson indices characterize the diversity of a community [76]. The alpha-diversity analysis based on OTUs, which is related to the Chao1 wealth analyser, is used to evaluate the abundance of microbial species or diversity of microbial organisms in different biogas-producing systems (Chao1); the abundance-based species-richness estimator (ACE), the Shannon–Weaver diversity index (Shannon) and the Simpson diversity index (Simpson) are used for the MOTHUR program, the VEGAN library software package [77], and the RDP Pipeline [61]. Of all the indexes, there is a positive correlation between Chao1, ACE, and Shannon and microbial culture biodiversity, although Simpson has negative correlation with it. In particular, the Shannon and Simpson indices can be used to assess biodiversity, while the Chao1 and ACE indices can be specifically utilised in the microbial communities to compute the total number of species (abundance). Therefore, the greater the Good’s distribution, the more the results of sequencing represent the real condition. One key benefit of utilising the presented equations to quickly estimate the biological diversity of microbial communities is that it allows for straightforward estimation, which is not the case with other diversity indices [78]. Simpson and Shannon indices combine richness and evenness into a single measure, obscuring other aspects such as the economic, ecological, and social value of each species [79]. Complex multivariate analyses such as canonical correlation analysis must be undertaken to alleviate the limits of biodiversity indexes and provide more important information on environmental variables connected with species in a microbial culture. In addition, the alpha rarefaction curve is used primarily to detect sequencing adequacy. It represents an association between sequencing numbers and species occurrence, which is typically developed using MOTHUR, QIIME, and R tools by randomly subsampling OTU sequences. The plateaus reflect adequate sequencing depth for alpha-rarefaction curves, while the monotonically rising portion is evidence of inadequate sequencing. For instance, Smith et al. [80] reported on a pilot-scale thermophilic anaerobic digester microbial culture system handling poultry faeces and contrasted the rarefaction curves for 16 S rRNA genes and various sequence similarities. Results proved that the rarefaction curve was asymptotic with similarity levels of about 97 percent, suggesting that the sequence dataset had adequately sampled variability in this study and that adequate sequence complexity at these levels was achieved. However, the amount of OTUs at 99–100% was still increasing, which implies much more undetected diversity at this stage [80].

3.4. Metataxonomic Analysis

Compositional analysis on the basis of taxonomy is the basic information that is obtained for anaerobic microbial community analysis. First, the database is compared and categorized, then, the sequences are taxonomically classified. Several sequence databases such as EzTaxon-e database [62], RDP database [81], SILVA database [82], and GenBank NT/NR database [83] are used to filter the unknown sequences through a blast search. The confidence level is generally 80–90%. Thereafter, using software such as Bergey’s taxonomy and Bayesian classifier [84,85], the resultant-matched sequences are allocated phylogenetically to the taxonomic classification. The analysis of taxonomic classification of anaerobic microbial populations is conveniently carried out on the basis of family, genus, order, class and phylum. This classification could be represented in a single multilevel species-taxonomy diagram using the software Krona [63]. Moreover, MetaPhlAn [86], PhyML [64], and CL community software [62], can also be used to find the phylogenetic composition of microbial populations. Zhang et al. [87] reported that the most dominant species were Trichococcus in first stage, i.e., hydrolysis, Amino bacterium in the second stage, i.e., acidogenesis, and Levilinea in the third stage, i.e., methanogenesis, in a self-designed three-stage anaerobic tank digesting food waste. Moreover, the dominant species of bacteria changed with a change in the feed stock found in the same anaerobic tank co-digesting horse manure and food waste. Archaeal species Methanosarcina and Methanobacterium and bacterial species Aminobacterium and Proteiniphilum were dominant species compared to mono-digesting species in the anaerobic tank [88]. Figure 5 shows a model diagram for the taxonomic hit distribution at the phylum level. The model graph shows the microbial community present in a digestate sample from an AD fed on agro-waste. In many studies, it has been found that the dominant species were altered with alterations in operating conditions such as pre-treatments [89] HRT, [47] OLR (organic loading rate) [57], temperature [90], etc, and type of feed stock. Metataxonomic analysis of a single and two-staged anaerobic digestion revealed that the species Clostridium, Bacteroides, Desulfovibrio, Lactobacillus, Lactococcus, Longilinea, Methanosaeta and Syntrophus showed changes in structure as changes in parameters such as measures of efficiency, hydrodynamics and kinetics of the performance were recorded [91]. In a recent study, a high-throughput sequencing method showed a shift in the archaeal and eubacterial population due to the inclusion of hydrogen gas in AD [92].

3.5. Study of the Similarities/Differences in Microbial Taxonomic Compositions

Based on the study of the composition of microbial taxonomy, study of the dissimilarity and similarity of different samples is carried out and the findings thus obtained can typically be reported in two ways, one being a statistical analysis plot and the other a ternary plot. Using a two-tailed Welch t-test with an alpha value of 0.05, STAMP [93] has also been used to conduct a pairwise statistical comparison of taxonomy between two samples. For example, Yu et al. [94] performed a comparative study between the relative abundance of traditional anaerobic digester (CAD) microbial communities and the “anaerobic dynamic membrane bioreactor” (ANDMBR) at the genus level, which resulted in a statistical analysis of differences. The definition of comparative excess is the fraction of sequences in a sample that are identical to a given taxon expressed as a percentage (percent). The findings suggest that the ADMBR includes a higher concentration of the genus Methanosarcina and a smaller volume of the genus Methanosaeta than the CAD in a statistically relevant way. Conversely, in the ANDMBR, Methanosarcina and Methanosaeta were more abundant as a whole than in the CAD [94]. Microbial taxonomic compositions in the various samples can be shown via a ternary plot [95]. To create a ternary plot in R use the ‘ternary plot’ function in the ‘ggtern’ 3.3.5 package [96,97]; depending on the relative existence of the plot, only the species present in all samples are stored and imported. The ratio of each species’ relative abundance to its overall abundance across all samples is used to describe the abundance of each species in each sample. Every species is distinguished by an exclusive point in the ternary sequence, and the proportions of the points correspond to the relative abundance. For example, in anaerobic granules from digesters fed with cannery wastewater, brewery wastewater, and milk wastewater, Lu [98] investigated the distribution of specific bacterial and archaeal populations using ternary plots and discovered that the majority of high-abundance bacterial phylae, such as Chloroflexi, Bacteroidetes, and Proteobacteria, were shared by all three types of granules. Around the same time, the phyla Methanosaeta occupied all cannery and brewery granules with more members of the phyla Methanobacterium found in cannery granules. The diversity and consistency of microbial communities between various anaerobic digesters can be distinguished using ternary plots. A model diagram depicting the taxonomic hit distribution is shown in Figure 6. In a metataxonomic analysis of an anaerobic digester fed with corncob, three different approaches were used to reveal that there is a noticeable change in the microbial consortium in the AD [99].

3.6. Statistical Multivariate Analysis

Through a brief literature survey concerning anaerobic microbial community analysis, the most commonly used multivariate-analysis techniques included “main coordinate analysis” (PCoA) [56,70,100], “non-metric multi-dimensional scaling” (NMDS) [49,101,102], “main component analysis” (PCA) [52,90], “redundancy analysis” (RDA) [36,48], and “canonical correspondence analysis” (CCA) [40,58,62]. All of these multivariate methods share the same goal of displaying comparable objects in close proximity to one another and dissimilar objects at a greater distance from one another [103]. Based on the data sets and computational methods employed, these methods are typically classified as either unrestricted (principal component analysis, principal component ordination, and NMDS) or constrained (ratio dissimilarity analysis, and canonical correlation analysis) (linear versus unimodal). An important distinction between PCA and PCoA in uncontrolled ordination work is that in PCoA the starting point vector is a parallel or dissimilar distance sequence, which is distinct from the PCA starting point for data collection. Nevertheless, in the study of complicated metagenome results, PCA, PCoA [104], QIIME [56], MOTHUR [70], and MG-RAST [17] have identified an increasing number of applications. For example, PCoA was used to compare populations of specific digesters using the same method of stabilization of biosolids and compared population differences between samples using different methods of stabilization [100]. In another case, PCoA studied the changes in the bacterial community structure in response to OLR, showing that the three-fold digesters behaved similarly at each point [40]. In comparison, with various OLRs, microbial population configurations shifted significantly from stage 1 to stage 4. Although there was negligible variation in community structure between stage 1 and stage 2, the enhanced OLR significantly altered the group structure in stage 3. The decreased OLR in stage 4 moved the bacterial population structure down to those in stage 1 and 2, but not absolutely [40]. Notably, it was more challenging for PCoA to evaluate variable participation as PCoA was unable to provide a direct link between the main components and environmental variables, so PCoA was preferred over PCA when a ton of input data was lacking [78]. Another unconstrained ordination technique, nonmetric multidimensional scaling (NMDS), is commonly used to describe the complexity and resemblance of microbial population systems in anaerobic digestion across time [48,105], or in several digesters under various environmental conditions, such as specific OLRs, temperatures [36,57], and additive (e.g., zeolites [106]) using the R package vegan [107]. In general, the NMDS technique produces the high dimensionality of the data by mapping comparable datasets together and mapping dissimilar datasets with larger stepped distances [103]. On the other hand, PCA is limited to using only the “Euclidean distance matrix”, but NMDS and PCoA are not limited to any particular distance matrix [108]. In general, however, NMDS is more computer-intensive than PCoA and PCA [103], resulting in a longer time needed for NMDS research. RDA and CCA are two commonly used methods for analysing metagenome data from microbial communities in restricted ordination science. These two approaches are very close, except that RDA is focused on linear models while CCA is based on unimodal interactions between organisms and climate [105]. For RDA plots a Hellinger transformation is performed, and the plots are created using vegan 2.5–7 package [97]. Next, a technique utilizing detrended communication analysis (DCA) can be conducted to help users choose which approach to follow [109]. The technique primarily involves four phases, i.e., (i) in the programme, selecting ‘DCA’ for indirect gradient analysis (e.g., CANOCO), (ii) choosing detrending by lines, (iii) choosing the Hill scaling for ordination-score scaling and (iv) selecting the other choices to run the study [109]. Through comparing the results of the measurements of gradient duration with 4.0 and 3.0, it is possible to quickly identify the most appropriate method. Specifically, unimodal strategies such as CCA should be used when the gradient duration value reaches 4.0; whereas the linear option is likely to be a better alternative if the value exceeds 3.0. RDA and CCA work reasonably well in the scale of 3 to 4 [109]. Because RDA and CCA are very close, RDA is given as an example by a brief case study. A CANOCO device (version 4.5, Wagenignen, The Netherlands) was used to conduct the RDA ordination. The bacterial community structure has been demonstrated to be affected by digester performance and operational factors [36]. The arrows reflect various environmental factors (e.g., operating conditions) in a standard CCA diagram, and the longer the ray is, the greater the environmental factor’s effect. A strong association between two environmental variables is seen where the angle between the environmental factors is extreme, while an obtuse angle corresponds to a negative correlation [108]. Reactor performance and operating parameters were examined, including temperature, pH, alkalinity, and volatile fatty acids, utilising CCA with XLSTAT methodologies to clarify the relationship between microbial (bacterial and archaeal) species and these factors (VFAs) [40,58,110]. The differences in sample periods among class-level microbial community composition and digester physicochemical characteristics is another great illustration of the anaerobic digestion method for the oxidation of organic industrial pollutants [111]. A Mantel test was performed using the integrated Mantel function in R, while the coordinates from a RDA plot of the material/inoculum combinations were transformed into distances using the geosphere 1.5–14 package (https://cran.r-project.org/web/packages/geosphere/geosphere.pdf, accessed on 13 November 2022) and Spearman method. The AD kinetic parameters were transformed into distances using the Euclidian method [97].

3.7. Metatranscriptomics and Application

Metagenomics data may be annotated dynamically utilizing bioinformatics frameworks or methods such as BLAST in databases (e.g., SWISSPROT [112], COG, KEGG [113]), MG-RAST (SEED) [65], GenDB [114], and IMG/M [115]) in order to obtain knowledge beyond taxonomic compositions. In particular, BLAST has been mainly used against various databases for the functional annotation of short-metagenome contiquity, while the GenDB genome-annotation software is usually responsible for the functional annotation of long-assembled contiquity [114]. The MG-RAST framework is a SEED-based program that allows users to access automatic analysis metadata [65]. IMG/M offers researchers a three-dimensional (genes, metabolites and functions) genome annotation system and comparative analysis of microbial genomes using several methods such as KEGG [116], InterPro [117], Pfam [118], and gene ontology [119]. Hybridization methods focused on microarrays are also helpful in the study of genome diversity and bacterial connectivity [120]. Although numerous bioinformatics methods for gene functional annotations have been developed, some of them, such as single genome genetic prediction tools, are not well suited for complex metagenomic data sets of anaerobic microbial populations due to heterogeneous sequence structure, length, and error [121,122]. It is clear that successful gene-classification methods (e.g., IMG/M and KEGG) can still be widely used to classify protein-coding areas, which is one of the main issues in microbial metagenomics studies. Such an example is the research by Campanaro et al. [65], which used bioinformatics methods (e.g., KEGG, COG and SEED) to model functional associations between microbial species involved in the AD process and to identify key microbial genomes that encode enzymes involved in different metabolic pathways. Microbial co-association networks include knowledge of the function of time or other external factors on the nature of population organization. Group metabolic networks, through recognizing metabolite exchanges and species-specific resource requirements, can provide a mechanistic connection between organisms [123]. In fact, MEGAN can also be used to systematically evaluate various metagenomes focused on the SEED hierarchy and KEGG pathways [68]. In reaction to minor disruptions, microbial populations may shift unexpectedly, related to changing conditions or several stable states. However, with time-varying networks, temporal variability in microbial encounters can be observed [124]. Therefore, more experiments under various operating conditions (e.g., specific substrates and temperatures) are likely to enable the anaerobic microbial genome database and related applications to be expanded in the near future through increasingly powerful tools and algorithms.

4. Application of the Analysis of Anaerobic Microbial Community on Biogas Production

Exploration in metagenomic data via bioinformatic tools has helped researchers to identify many important bacterial and archaeal species in anaerobic digester tanks [17,65,125]. Studies have also revealed many different parameters such as, changes in microbial population according to the alteration of operational parameters, methanogenesis pathways and bacterial metabolism [83,89], and performance of anaerobic digesters with respect to the microbial community [126,127,128]. Furthermore, technologies are being established in the field of metagenomics using new and improved bioinformatics tools, helping researchers to gain knowledge about new developments in the field of biogas production. Various developments, viz., improved and optimized AD processes, pliability and rigidity of metabolic pathways, analysis of antibiotic-resistant genes and mathematical modelling of microbial communities are summarized.

4.1. Development of Improved AD Processes

The optimization and improvisation of AD are achieved by improving various strategies such as pre-treatment of feedstock [129], optimization of operational parameters [130], design optimization of the anaerobic digester [87,88], addition of supplements such as additives [131], etc., and microbial community analyses are mostly carried out for all the alterations conducted for optimization of AD. Zhang et al. [131] reported that the process of anaerobic digestion in both pilot-scale as well as lab-scale digesters was improved by adding activated carbon as a supplement. 16 S rRNA genes were pyrosequenced to find that the dominant phyla, i.e., Proteobacteria, Elusimicrobia and Firmicutes, increased by 1.7–2.9 times due to supplementing the digesters with activated carbon. Another study conducted on hydrogenase and nitrate reductase at the transcript level revealed that Blautia, Acetitomaculum (acetogens), Selenomonas, Wolinella (fumarate and nitrate reducers) and Desulfovibrio (sulfate reducers) were the main cause for H₂ production in AD [132].

Recent research revealed that in three-staged digesters, all three stages, that is hydrolysis, acidogenesis and methanogenesis, were optimized and selectively enriched at every stage with respect to the microbial community of each stage, i.e., hydrolysing bacteria, acidogenesis bacteria, and methanogens. That study was conducted using pyrosequencing analysis, and 16 S rRNA high-throughput sequencing is yet to be performed [87,88]. In other research Suksong et al. [130] revealed that the highest methane yield was obtained at a carbon to nitrogen (C:N) ratio of 30:1, total solid (TS) content of 16%, and feedstock to inoculum (F:I) ratio of 2:1. This inference was obtained by optimizing the operational parameters such as TS, C:N ratio, and F:I ratio. The enhanced production of methane gas was due to the selectively enriched microbial community, viz., Clostridium species, Methanoculleus species, and Ruminococcus species [130]. A recent study revealed that despite producing methane at a rate of 241–247 mL/g VS, the thermal hydrolysis process-anaerobic digester (THP-AD) reactors collected around ~1999 mg/L of propionate. About 30 significant “metagenomic assembled genomes” were augmented and ~70 metagenome-assembled genomes were retrieved from that metagenomic study. A study of genome-centric metagenomics revealed the recovery of 68 metagenome-assembled genomes (MAGs), 32 of which were significantly enriched [133]. A study of the effect of carrier-based silica-lignin on anaerobic digesters showed that maximum bacterial count was noted during dehydrogenase phase [134]. The overall improvement of AD can be achieved by study of genome–centric analyses for understanding the whole microbial community. This could be accomplished by developing an assembly of whole functionality of the microbial community and assessing its coding gene annotations [135].

4.2. Development on Metabolomics Analysis

Treatment of unmanaged waste and production of biomethane gas can be accomplished with the help of highly efficient anaerobic reactions. However, the efficiency of the anaerobic digestion processes could be increased by gaining sufficient knowledge about the metabolic pathways of methanogens. Recent studies have mentioned the coupling of metagenomic sequence analysis along with radio-isotopic analysis to disclose the main metabolic pathways of dominant methanogen species present in the anaerobic tanks, digesting manure or sludge [17]. Results from the two different analysis methods have shown inconsistencies, i.e., the metabolic pathways revealed by radio-isotope analysis were different from the results determined by metagenomic analysis. Many similar studies have been carried out on networks of microbial communities and their metabolic pathways with the help of several bioinformatic tools and techniques. Likewise, research on metabolic pathways and networks of microbial communities using bioinformatics technology are available in automated AD systems. Zhang et al. [18] reported that a food-waste-digesting anaerobic digester was optimized by enriching it with activated carbon. Analysis of metabolic pathway with the help of KEGG-pathway analysis revealed that there were significant changes in the microbial community and their metabolic pathways, which was also confirmed by taxonomic tree analysis. The results of both analyses showed that the main metabolic pathway in ADs enriched with activated carbon was the metabolic pathway of propanoate which transforms propanoic acid into acetic acid. Lipid metabolism and methanogenesis pathways were also improved and it was found through the microbial community analysis that the activated carbon encouraged the growth of the species Methanosaeta and Geobacter, which form a highly populated syntrophic microbial community [18]. Jiang et al. [136] recently studied the mcrA (methyl Co-A reductase) gene and 14C-labelled sodium acetate by using a pyrosequencing technique and radiolabelling technique to find out how the relationship between dominating methanogen pathways and TAN (total ammonia nitrogen) is altered by change in syntrophic acetate oxidation and acetoclastic pathways, which further results in enhanced methane formation in AD. It was also found that the higher carbon dioxide and methane ratios (2.1–3.0) in high total ammonia nitrogen (TAN), approx. 11.1 g/kg wet weight, digesters show a maximum of 75% methane production through hydrogenotrophic pathways, whereas, only 23% methane production was through hydrogenotrophic pathways when the ratio of CO₂ and CH₄ was approx. 0.1–0.3 in low TAN (0.2 g/kg wet weight) digesters [136]. Thus, the relation between TAN and metabolic pathways could be optimized to manipulate the syntrophic and acetoclastic metabolic pathways. In another study of continuous AD, a 16 S rRNA gene-sequencing technique found that there was a huge change in the microbial community during its adjustment to an environment with a high concentration of ammonia (10 g NH⁴⁺-N/L) [137]. A hydrogenotrophic species, Methanoculleus spp. and a syntrophic acetate-oxidizing bacterial species, Clostridium ultunense, were found to be growing rapidly and increasing in population at high levels of ammonia (>7 g NH⁴⁺-N/L), revealing a hydrogenotrophic pathway as a major metabolic pathway [137]. A major shift from acetoclastic pathways to more flexible and complex metabolic pathways was seen at 5 g/L NH⁴⁺-N concentration [138]. In another study, changes in the microbial community due to change in temperature was found. Metabolic pathways were studied in anaerobic tanks digesting activated sludge, working at thermophilic temperature range (55–65 °C) and 2–4 days of HRT (hydraulic retention time) [83]. At higher temperature, i.e., 55–60 °C, Methanosarcina species were found to be dominant as revealed by 16S rRNA pyrosequencing analysis. However, the analysis also revealed that if the temperature is raised to 65 °C, the population of Methanosarcina species is significantly reduced, resulting in a decrease in overall methane yield and deposition of VFA. Higher temperature encouraged the syntrophic acetate oxidation as found by stable-isotopic signature (δ13C) analysis. Similar results were seen when HRT was reduced to two days [52]. In a recent study, raw sequence data were subjected to quality assurance, splitting, and grouping in order to create OTUs, and then species taxonomic classification was conducted. “Canoco5” was used to conduct the analysis for redundancy. This study revealed that feedstock breakdown sped up as a result of co-digestion, which enriched the dominating hydrolytic bacterium Defluviitoga [139]. Formation of phosphine in an anaerobic digester was investigated in a study which revealed that the bacterial species Escherichia and Ruminococcaceae were found in abundance. Two types of databases: KEGG and MetaCyc pathway revealed the close relation of phosphine synthesis [140]. Bioaugmentation in anaerobic digesters helps in recovery of VFA, and a metabolomics study revealed that during the early steps of bioaugmentation of VFA, the acetoclastic microbial population plays the major role in methanogenesis and neutralizing the pH [141]. In a recent study of a multi-staged anaerobic digester fed with food waste, an analysis was conducted on the foaming mechanism in the digester. The results of the metabolomic analysis revealed that the metabolic molecules that have higher expression, are up-regulated, and have high surface activity are the major contributors in foaming [141]. Metabolomic analysis during co-digestion of corn straw and pig manure showed the hydrogenotrophic and acetoclastic pathways working during the process [142]. Another study using beta-diversity analysis revealed that richness in species is more significant in thermophilic digesters than in mesophilic digesters [143].

4.3. Development in Identification of Antibiotic-Resistant Genes

Metagenomics methods, bioinformatic tools and techniques, and high-throughput sequencing are successfully used to identify antibiotic-resistant genes as disposal of the sludge or digestate in open environments, specially on land, may cause health risks. Lee et al. [144] presented in their study a characterization of wastes such as. sludge, food waste and manure in terms of amount of antibiotic-resistant genes. The result of the characterization revealed that comparative amounts of antibiotic-resistant genes were found to be highest in manure and the least amount was found in food waste. The results showed that the diversity and structures of antibiotic-resistant genes varies with variation in the type of substrate [144]. In a recent report, Luo et al. [126] reported a significant overall amount of antibiotic-resistant genes, 7 × 10 ⁻³ to 1.1 × 10 ⁻¹ copies of ARGs/copy of 16 S rRNA DNA, in tanks digesting, industrial waste and manure. The digestate from thermophilic anaerobic digesters has a lower amount of antibiotic-resistant genes [126], which makes thermophilic digesters more dominant for removing pathogenic microorganisms [145,146]. Antibiotic-resistant genes and bacteria resistant to antibiotics were observed in ADs digesting pig manure through HTFQ PCR (high-throughput fluorescent quantitative PCR) and Illumina MiSeq sequencing [147]. Three transposons and 83 antibiotic-resistant genes were detected from the above-mentioned sequencing techniques. Also, some resistant genes such as tetracycline and Macrolide-lincosamide-streptogramin decreased after the AD processes. However, the comparatively large quantities of florfenicol, sulfa drug, aminoglycoside, chloramphenicol, and amphenicol-resistant genes were improved by 52–270 times [147]. Biogas digestate has been used as manure for farmlands for a long period of time and the presence of antibiotic-resistant genes has created an increment of the population of antibiotic-resistant bacterial strains causing contamination of cropland soil [147]. Therefore, to minimize the proliferation of antibiotic-resistant genes in farmland soil, remedial and regulatory steps must be employed. Recent research has revealed that pre-treatment of substrate, i.e., sewage sludge with microwaves, helped in reducing the antibiotic-resistant genes, which, however, increased in the AD but in nominal amounts [110]. Hence, pre-treatment with microwaves is one of the easy and potent techniques for removing antibiotic-resistant genes from bio-manure made from AD digestate. To better understand antibiotic-resistant and bacterial species, environmental metagenomics analysis techniques have been explored and developed; for example, an online open analysis portal, ARGs-OAP is one such platform which consists of a databank named SARG version 2.0 [148]. The SARG databank consists of a lot of sequence data present in ARDB, CARD, and protein databanks, namely NCBI-NR [148]. For easy and complete access to data from the ARGs-OAP databank online portal, http://smile.hku.hk/SARGs (accessed on 13 November 2022) can be browsed.

4.4. Creation of the Modelling and Optimization of Microbial Population Dynamics

Some recent trends have revealed many developments in microbial community-dynamics modelling and how these dynamics are related to operational and physiochemical parameters. For example, in order to promote rational control and intervention in microbial communities, Hanemaaijer et al. [149] determined analytical modelling methods for microbial community dynamics, ranging from methagenomics to community structure. The study was dedicated to two kind of modelling methods including a naïve coarse-grained model which could resolve the extrapolating problem from the investigational information, and the other model was a mathematical model which could assimilate prevailing physiochemical, physiological, and genomic data [149]. The first model has established unusually low consideration, although it can possibly be improved to be a more complete genomic-level stoichiometric model which can perform as mixed data integrators. The second model has been primarily considered and used because of its ability to exploit data and extrapolate power [149]. Similarly, Succurro et al. [150] gave a summary of mathematical models of microbial ecosystems found naturally and accentuated that to develop a hypothetical account of a microbial community one should choose an exact problem. A current agreement is that it is time for more progress to be made in the large theoretically defined methods, the methagenomic microbial interface simulators (e.g., MetaMIS [151]) and the mathematical amalgamation tools (e.g., MetaTopics [152]) to expand the population dynamics area of modelling from a mere explanatory example to a whole biogeochemical concept [150,153,154]. Further, Succurro and Ebenhöh [155] published a detailed review of mathematical modelling of microbial ecosystems, which determined two main divisions of mathematical approaches (differential equation models and constraint-based stoichiometric models) and presented the current purpose to the analysis of microbial communities. The benefits of these two models are determined tactics that can effectively provide a macroscopic illustration of microscopic biological systems [155]. In order to move forward, an integration modelling method has been shown to recognize growing patterns in microbial classification and their dynamics delimited by diverse spatial-temporal phenomena [155]. It was proposed that simulators and researchers would work together from the theoretical phases of the investigation plan to guarantee a precise mixing of concepts and experiments [150]. With the help of mathematical models, a common platform could be made to understand the structures of complicated bacterial communications. The results and information attained from the modelling of microbial populations help in controlling and enhancing the AD process by optimizing the microorganisms. In this regard, many significant advancements in microbial population systems alterations have been made. Digester conformation, type of feedstock, environment conditions and operating factors (e.g., temperature) are the important pushing aspects for community-structure deviations [156]. For example, in an anaerobic tank digesting pig excreta, the genus Methanocorpusculum was in the majority at pH 7.0, and at pH 6.0 and 8.0, genus Methanosarcina was dominant, showing that by controlling pH dominant species of microbial populations can be optimized [157]. Likewise, some research has been conducted to produce a feedback system which can efficiently help in detection of any malfunction in the overall system. For example, alkalinity is used as a sign or indicator for steadiness of the process in an AD [158,159,160]. So far these technologies have brought breakthrough advances which have helped in better understanding AD processes, however imaging the multifaceted 3-D multiple species population nevertheless creates substantial challenges in experiments, so the knowledge gained so far and utilized for optimizing and modifying existing AD processes needs to be researched further.

5. Restrictions and Predictions of Bioinformatics Analysis on Microbial Metagenomics

Analysis of the microbial community, i.e., metagenomic analysis of anaerobic digesters has been frequently explored. Most countries are working in the field of metagenomics and meta-transcriptome analysis and microbial community analysis of anaerobic digesters. However, there are many challenges that need to be tackled before bioinformatic analysis becomes a general technique. Advancement in sequencing technology has made the sequencing process rapid and easy and therefore has caused some serious problems in data storage. For example, a huge amount of data, of about 600–1000 GB per run is being produced by sequencing data through Hi Seq. 2500, and similarly approx. 1,000,000 reads are produced by 454 pyrosequencing on each sequencing run, which is equal to about 0.7 GB data [146,161]. Again, raw sequences increase the sequencing data up to 10–20 times for each analysis [161]. A major challenge in the field of bioinformatics is the enormous volume of data which is being produced due to sequencing [122]. Furthermore, metagenomic research based on next-generation sequencing is still naïve. Recently sequenced metagenome data should be provided with more storage space and statistical sustenance. Therefore, to explore microbial communities of anaerobic digestion a common program or platform for the storage of enormous data must be created. Also, statistical methods, machine learning, artificial intelligence, and data mining are some of the fields from which the tools and techniques of bioinformatics have been derived for data management [162]. However, the original data of metagenomics which are generally fed into bioinformatic software differ greatly with the genomic data [162]. Genomic data analysis has been carried out using various computational approaches; however, there are some issues that need to be resolved. Humongous sequence data is an insuperable problem for assembly and binning which can be accredited to the incomplete calculating ability and explicit internal storage ability [39]. Moreover, there occurs an absurdity between precision and data managing rates because assembly and binning according to the taxonomy requires numerous hours to numerous days [161]. Therefore, more effective analysis tools should be developed that are rapid in completing analysis tasks such as the technology present in super computers and also provide more precise results. Another reason that has been found to be an obstacle in metagenomic analysis is the excessively high amount of cost of metagenomic sequencing, consequently, repeatability of experiments become less feasible, and hence it is difficult to confirm the significance of produced data [33]. Thus, to obtain more consistent and unfailing results, it is advisable to sequence in triplicate. Campanaro et al. [163] in their study stated that biasness of sequenced data of amplified 16S rRNA genes is mainly because of the improper matching of the universal primers of all the target genes and also because of lesser-known hypervariable regions [163]. The usage of several indicator genes and transcriptional studies are therefore suggested in order to improve the estimation of profusion for important taxonomic groups [163]. In fact, most bioinformatics analyses to date have relied on the contrast of sequences with reference databases. Practices for using existing databases for evaluating bioinformatics, however, may be troublesome given the potential incompleteness of datasets. Therefore, to promote reliable and efficient study of anaerobic microbial metagenomics, upgraded databases with better quality data are required. Fourth, metagenomics strategies provide a deeper insight into microbial populations, but have primarily been used for concise and informative methods to react to the original question [122]. There are still differences between research, pilot and field data, and extrapolation. In addition, there is still a substantial gap in what we can know from microbial communities through bioinformatics methods and what species and gene functions we can regulate, because the microbial environment is an extremely complex network of diverse spatio-temporal interactions between microorganisms as well as between microorganisms and the ecosystem [164]. In order to fill this void, further information should be offered on how the genetic properties produce their dynamic behaviours and the dynamic interactions between individual microorganisms. In fact, nutrient amounts and fluxes within microbial species should also be thoroughly investigated. Finally, the existing scope of mining of anaerobic microorganism metagenomics data, based primarily on clear sequence similarity searches, may not be sufficient as microbial genomes should be the basis for microbial ecology [165]. For more comprehensive data mining, it is highly recommended that methodologies focused on bioinformatics be combined with core biotechnological approaches (e.g., TRFLP, FISH, PCR-DGGE and qPCR [1,27,166]). Nevertheless, there may still be a long way to go to the ideal point where digester efficiency can be precisely managed on the basis of anaerobic microbial community real-time analysis of methagenomics information. A large-scale precision-fermentation program using an artificial neural network is proposed to accomplish these efforts and to continue with the production of insightful solutions. With an artificial intelligence (AI) workflow, vast data (e.g., elementary feedstock materials, operating parameters, and microbial population metagenomics) derived from extensive past AD activities can be built into coevolutionary neural networks that can be further refined into useful networks such as deep Q networks using experimental data as training details. A better combination of regulatory parameters, digester sorting, and microbial population information can be proposed in a useful network with new feedstock part data to support practical AD operations. Although there are merely a few research reports [167,168,169] built on the disposition of AI methods over the last years, AI application in anaerobic digestion simulation, regularization, and optimization has great potential in the near future, particularly with the rapid development of novel and hybrid AI approaches. For example, a new algorithm has been created to design minimal microbial communities with required metabolic capability [170]. This algorithm will precisely enable researchers to identify minimum sets of microbial species that can collectively provide the enzymatic capability needed to synthesize the target metabolite population from a predefined selection of accessible substrates [170]. Thus, the result might be a major move towards the objective of modifying specific microbial communities for the optimization and industrial applications of anaerobic digesters.

6. Conclusions

Lately, a lot of research work has been conducted in the field of microbial community analysis and metagenomics analysis of anaerobic digesters that has helped to better understand the AD process. New techniques and tools in bioinformatics have helped in controlling and optimizing the anaerobic digestion processes in terms of feedstock, temperature, pH, HRT, VFA, TAN, etc., which has been a boon for the industries and has huge commercial potential. Despite the many applications of metagenomic analysis, there are some major issues related to this analysis such as data consistency, data storage, data usage and data managing techniques. These problems need to be assessed and tackled to further improvise and control AD processes. This review gives insights about the latest trends in metagenomic studies, newest and often used bioinformatic tools and techniques. However, the prevailing issues in these type of studies have grabbed the attention of researchers from the fields of environmental science, biotechnology, bioinformatics, computer science, and civil and chemical engineering. A lot of demand for research and development in this area is the need of hour and to strategically implement new advancements to provide molecular techniques at the commercial level. New technology in the field of database management is now being researched and technology such as artificial neural network and machine learning are finding their way into the optimization of biogas producing anaerobic tanks. Henceforth, the near future is brightening up with usage of more advanced technology in bioinformatics and to produce better yields of methane through optimally engineered anaerobic digesters.