Recent rapid advances in sequencing technologies provided a relatively cheap and fast way of studying the diversity of microbial species by discovering representatives of novel divisions or even phyla [5,6] and analyzing the variation within the species by sequencing closely related genomes from the ecological microbial populations or clinical studies of pathogenic bacteria.

Historically, prokaryotic organisms were organized by classical taxonomic ranking system (species, genus, family, order, and phylum). Delineation of prokaryotic species was originally based on phenotypic information, pathogenicity and environmental observations. A recent review [7] describes the history and present state of various methods of description of prokaryotic species. The authors suggest the concept of species as “a category that circumscribes monophyletic, and genomically and phenotypically coherent populations of individuals that can be clearly discriminated from other such entities by means of standardized parameters”. Scientists from different disciplines (taxonomists, ecologists and evolutionary biologists) have different interpretations of species defined by the framework of their needs and the tools they use for identification. For almost 50 years DNADNA Hybridization (DDH) has been the gold standard method for prokaryotic species delineation at the genomic level. New approaches to the identification of microbial species are taking into account the advantages of the growing massive volume of genomic sequence data [8,9]. Several groups have attempted to delineate the taxonomy of Archaea and Bacteria using the methods based on single-copy universally conserved gene markers [10-14]. Other method employ genome sequence directly without pre-annotation of genes. Average Nucleotide Index (ANI) is an silico version of DDH method. JSpecies package [15] provides an interface to calculate ANI for a pair of genomes that can be used for species classification. The ANIbased genome tree can be used to model the organism evolutionary relationships. Genome Blast Distance Phylogeny (GDBP) method [16] is another implementation of in silico DDH method that infers genome-to-genome distances between genome pairs from genome sequences. K-mer based approach uses the number of co-occurring k-mers (substrings of K nucleotide in genomic sequence) as the distance between the pair of genomes that represent the evolutionary relatedness [17]. The benchmarking of genome based classification methods can be found in the recent review [18].

It is now known that intra species variation can be as significant as interspecies diversity. Bacterial genomes from various strains of the same species can vary considerably in genome size, nucleotide composition and gene content. It has become clear that, bacterial species cannot be represented by an individual reference strain or a set of reference genomes. The ‘pan-genome’ concept has been introduced by Tettelin et al. in 2005 [19]. The pan-genome has been defined as a super-set of all genes in all the strains of a species. A pan-genome includes the “core genes” that are present in nearly all strains; “accessory genes” present in two or more strains, and finally “unique genes” specific to single strains. A shift in the paradigm from individual genome to ‘pan-genome’ has occurred in the past few years, with the rapid advances in the sequencing technology. The main approach in pan-genome studies is a comparative analysis of multiple strains from a single species, although one can also describe pan-genomes for different taxonomy levels – for example, a phylum or genus pan-genome, or sometimes even a subspecies pan-genome (as in the case of E. coli O157:H7, with 34 genomes sequenced so far). Alternative approaches include the use of rapidly growing metagenomic sequence data and single-cell genome sequencing. The pangenome concept is already changing the way we understand bacterial evolution, adaptation, and population structure, and has further important implications in identification of virulence genes [20].

A recent shift in the paradigm changes the field of comparative and population genomics. This emerging field leads to the development of new algorithms, statistical models, and visualization tools. It is affecting all areas of bioinformatics including data storage and data management, genome assembly and annotation, protein clustering, phylogenetic trees construction.

Over the last 20 years, the National Center for Biotechnology Information (NCBI), as a primary public repository of genomic sequence data, has been collecting and maintaining humongous amounts of heterogeneous data [21]. The databases vary in size, data types, design and implementation. They cover most of the genomic biology data types including the project description, project sequence data (genomic, transcript, protein sequences), sequence reads and related bibliographical data. All these databases are integrated in a single Entrez system and use a common engine for data search and retrieval. This provides researchers with a common interface and simplifies navigation through the large information space.

The Genome database was first created in 1995 when the complete genome of Haemophilus influenza, the first cellular organism was sequenced at TIGR [22] and submitted to GenBank. The genome data has changed dramatically over the past 20 years of microbial genome sequencing [23]. The newly redesigned Genome resource organizes information about the organism (usually at the species level) and provides the summary view of the data from all “genome-scale” projects: map, genome, assembly, annotation, transcriptome etc. The daily updated list of all genomes can be browsed via web (https://www. ncbi.nlm.nih.gov/genome/browse/) or downloaded via FTP (ftp://ftp. ncbi.nlm.nih.gov/genomes/).

NCBI, as a primary data archive, traditionally collected species and strain designation from the submitters of the sequence data. In the absence of the independent validation methods that leads to potential problems with the taxonomic identification. One of the most premises when classifying new taxa is the designation of one of the strains as being the type material that should be used as reference for any further taxonomic work. NCBI Taxonomy [24] team has started a project to curate type material in the taxonomy, and use it to flag sequence from type in GenBank. NCBI is developing new approaches to the classification of genome sequence and correction of sequence tags in GenBank. They employ all three described above computational methods (marker-based, K-mer, ANI) to construct the genome distance trees and evaluate the species assignment. Sequences from type can use those as landmarks of correct identification to help resolve misidentified genomes.

The data in microbial genomes collection have different levels of sequence and assembly quality; some sequencing technologies have known high level of error rate that lead to the large number of ambiguities and low quality nucleotides in the final consensus sequence. Many genomes assemblies coming from single cell sequencing technology give only partial representation of DNA in a cell, ranging from 10% to 90%. NCBI Refseq project aims to provide a high quality data set of genome, transcript and protein sequence data. The source of the genomic sequence in the Refseq collection is a primary sequence record in the International Nucleotide Sequence Database Consortium, INSDC, public archives [25].All assemblies with full representation of the genome of a single organism that pass validation quality are taken into Refseq. Genome assembly quality validation criteria are described in details in [26-28]. Genome assemblies from environmental samples, mixed cultures, hybrid organisms and chimeras submitted to GenBank are not accepted into RefSeq because they do not represent a single bacterial organism.

Integrated microbial genome resource provides an infrastructure for comparative genome analysis. Microbial sequencing projects now span from complete and draft genome assemblies of isolated organisms to large-scale comparative genomic projects of multiple strains, and to the new field of metagenomics where the entire complement of DNA from a given ecological niche is being sequenced. The information form tens of thousands of sequenced genomes has already provided an insight into microbial diversity, evolution, and ecology. Advances in sequencing technologies continue to change the field of genomics creating great opportunities of developing new bioinformatics approaches and computational sequence analysis methods.