Comparing pan-genomes amounts to comparing two graphs, generalizing the idea to align two genomes. We aim at developing algorithms and software for ‘whole-pan-genome alignment’. Though for aligning two networks approaches already exist, they do not address the peculiarities of pan-genome graphs, in, for example, de Bruijn graph or variation graph based pan-genomic data structures. We will address these particular issues in full detail. In addition, we will study phylogenetic relations within and across pan-genomes. In applications, in collaboration with INRIA Grenoble Rhône-Alpes, we will compare also ‘pan-transcriptomes’, e.g. liver cell with colon pan-transcriptome, and microbiomes, integrated into pan-genomes, so as to accurately identify the commonalities and differences between the two entities.
1st year: courses and training; for research we expect to provide accurate models for pan-genomic comparison with lots of applications to test them on; and first results on faster algorithms for performing graph/network alignment problems.
2nd year: for transcriptomics we expect to exploit further structure for faster running times; model phylogenetic problems as they arise in pan-genomics.
3rd year: design phylogenetic algorithms and write the PhD-thesis.