K-GCAK: K-mer-based Genome Composition Analysis Knowledgebase

Genome composition, concerning the variability of the four nucleotides (T, C, A and G) and even their translational products of 20 amino acids, is fundamental genomic and genetic information. We mathematically define nucleotide and amino acid compositions in DNA and protein sequences, respectively, in fixed lengths (k-mers) and establish a database to host the processed data derived from k-mer arrays (k=2-10 for DNA and k=1-5 for protein) for a large number of genomes. We collect both nuclear and organellar genomes as well as their annotations from multiple resources (virus, prokaryotes, protists, animals, and plants) and divide genes into protein-coding and non-coding groups. The content of this database is partitioned into three levels: species name, sequence type, and the size of k-mers. We provide estimated statistical results of different k-mers for various elements in genomes including k-mer parameters and genome complexity measures (information entropy and distance to even distribution). We also offer pre-computed figures of k-mer distribution and tendency of uniqueness ratio changes by means of Jellyfish and Tallymer, respectively. We identify unique sequence signatures with both options of 5’-preference and 3’-preference to collect the potential primer sequences in selected eukaryotic genomes, aiding the design of primers in PCR and probes in microarray experiments. In addition, we introduce four operation functionalities using the data involved in K-GCAK on genomic complexity, Jellyfish, Tallymer, and phylogenetic tree construction using k-mer (distance between species’ k-mer arrays), and users can choose their interested multiple species and genomic elements to compare their genomic complexity, saturation value of uniqueness ratio, and topological/evolutionary relationship among them. Using K-GCAK, we hope to address several important theoretical and bioinformatic questions across the same elements in different species or different elements in the same species. For instance, to understand the link between genome components and organism phenotype, to determine species-specific parameter settings in sequence assembly process, and to build phylogenies using different genomic elements. K-GCAK is freely accessible at kgcak.big.ac.cn/KGCAK/.