Making sense of genomes

Sequencing produces enormous amounts of data. Computational genomics uses high-performance computing to assemble genomes, annotate them (work out what each part does), and compare them. Common tools include BLAST (to search for similar sequences), ClustalW (to line up several sequences) and Phylip (to build family trees of species). Graphics processors (GPUs) speed up the work.

In India, the Department of Biotechnology has set up three National Genomics Core facilities: at CDFD, Hyderabad; NIBMG, Kalyani; and the University of Allahabad, Prayagraj.

Genome studies fall into three branches:

• Structural genomics: working out the three-dimensional structure of every protein coded in a genome (and finding new protein shapes, or folds); or studying how the genome itself is physically organised, from nucleosomes to large rearrangements. For example, mice and humans share many genes, but they lie on different chromosomes.
• Functional genomics: how the information in the genome is put to work in a cell, in normal and diseased states. It uses methods such as RNA-Seq, ChIP-Seq and metagenomics.
• Comparative genomics: comparing genomes of different species or individuals. It helps annotate new genomes, find the core genome (genes all members share) and genes unique to a species, and build taxonomy and evolutionary trees based on whole genomes.