Research Overview
Computational Biology/Bioinformatics is
a rapidly growing field of research that is constantly being
transformed by the availability of new technologies to probe biological
molecules and systems. Quite often these technologies can
produce vast amounts of information with a fair amount of
noise and researchers have to rely on computational tools to sift
through the data to make sense out of it. Biological systems are
however notoriously hard to model and the compulsions of processing
large datasets sometimes leads to the adoption of heuristic solutions
to these problems. In my experience, however, even surprisingly
innocent heuristics/approximations can lead to unexpected
results and fundamentally affect the biological conclusions from a
dataset (see also:
Motif
Finding,
Computational
Statistics).
In many contexts, heuristics can be unavoidable due to the
computational complexity of the problem (see also:
Genome Assembly) --
even then, it makes sense to study the problem under a formal framework
to characterize the limits of both heuristics and more principled
solutions. My personal take on this is that: "Data is expensive and
hard to generate -- computational resources are relatively cheap. Its
therefore worth putting
extra effort into more well-founded and precise computational analysis
to get the most out of the data".
In terms of areas of research, my work
in Computational Biology mostly falls under two broad categories:
Genome Assembly (the
task of computationally reconstructing the genome from experimental
data) and
Motif finding
(finding functional motifs in biological sequences). In my recent work,
I have also looked at computational and statistical issues in
Metagenomics (studying
uncultured microbial and viral samples) and some aspects of
Phylogenetics
(reconstructing evolutionary histories of sequences and organisms).
More information on these topics can be found in the links below:
