Title: Single-cell Genome Sequencing and Assembly: Progress and Prospects

Abstract:

Recent technological advances in high throughput low-cost DNA sequencing have brought a whole new realm of exciting applications within reach, one of which is genomic analysis at single-cell resolution. Single-cell genome sequencing holds great promise for various areas of biology. The vast majority of environmental bacteria, e.g. from human microbiome, cannot be cultivated in isolation as they require their symbiotic natural habitat to grow. Also, a cancerous tumor is heterogeneous as the constituent cells quickly accumulate random mutations and aberrations in the absence of tumor suppressor genes, and an important problem in cancer genomics is discovery of the first few initial variations that start the tumor. A solution to that problem consists in reconstruction of the tumor phylogeny from the genomes of single cells sampled from different parts of the tumor. Our ability to acquire and analyze genomic sequences at single-cell resolution is expected to have significant impact on energy, environmental, and health research. Single-cell sequencing is currently a challenging task since it relies on whole genome amplification techniques which introduce coverage bias, chimeric amplicons, and sequence errors.

In this talk, we will first present our single-cell assembly algorithms, Velvet-SC and HyDA, devised to mitigate single-cell artifacts. We will then present our algorithm on synergistic colored co-assembly of multiple single-cell data sets. Our results suggest that the colored co-assembly significantly improves assembly quality and enables discovery of some interactions between the input species that would not be possible using individual single-cell assemblers. We will then present our divide-and-conquer compressive sensing algorithm to capture all the distinct genomes in a sample, exploiting the fact that the number of distinct genomes is much less (e.g. ~10^3 in human gut microbiome) than the number of cells (e.g. ~10^10). Finally, we will mention a few future directions to pursue in the emerging subfield of single-cell genomics.