Biomarker Discovery via NGS-based Transcriptomics: A Case Study
Kshitish Acharya, Faculty Scientist, IBAB
One of the key reasons for slower
rate of discovery of new diagnostic or therapeutic molecules, in recent years,
is lack of attention to thorough screening and identification of target
molecules. If researchers do not identify the right target molecule it can
result in failure at later stages of R & D or during clinical trials.
Exploring gene expression profiles is key to biomarker
discovery, which in turn is crucial for diagnostics, prognostics and
therapeutics. With the advent of modern techniques such as the microarrays, NGS
and mass spectrophotometry, and parallel bio-IT approaches, there is a higher
potential to generate more data and a more meaningful short-listing of
potential target molecules. RNA-sequencing particularly offers hope to identify
key transcript- and/or protein-isoforms associated with diseases. But the data
need to be carefully analyzed and interpreted. This has been a non-obvious challenge.
There seems to be a tremendous gap between general biologists, health
professionals, molecular biologists, pharmacologists and computational
biologists/bioinformaticians. We have been doing some research in meeting such
challenges, both at IBAB (www.ibab.ac.in) and Shodhaka (www.shodhaka.com). We earlier
developed a simple yet effective computational meta analysis method. We also carefully
compiled public transcriptomic data and developed a few software and databases
for better analysis and interpretation of the data. Using our new methods, we
identified a list of genes and specific alternatively spliced forms of
transcripts, which may be important for a type of male infertility
(non-obstructive azoospermia - which we think is a good model to work towards
better male contraceptives as well). We then performed our own RNA-sequencing
using clinical samples, and validated the observations. We are currently
performing more data analysis, particularly in terms of network analysis, and
pick crucial biomarker-candidates for non-obstructive azoospermia as well as a
few other disease conditions, such as lung and breast cancer, as well.
|