I work with Dr. Peter Woolf in developing machine learning methods to learn mechanisms from biological data and knowledge. My methods and software use Bayesian networks to simulate complex biological systems using probabilistic and nonlinear models.
I am completing my dissertation tentatively titled Mechanistic Bayesian Networks for Biological Modeling and expect to graduate in August 2009.
Shah A, Keshamouni V, Woolf P.
Validating protein and miRNA interaction networks using
expression data.
In Preparation, 2009.
Shah A, Tenzen T, McMahon A, Woolf P.
Using Mechanistic Bayesian Networks to Identify the
Downstream Targets of Sonic Hedgehog Pathway.
Submitted, 2009.
Shah A, Woolf P.
Python Environment for Bayesian Learning: Inferring the
Structure of Bayesian Networks from Knowledge and
Data.
Journal of Machine Learning Research.
10(Feb):159-162, 2009.
abstract /
pdf /
bibtex /
RIS /
mloss.org
Draghici S, Khatri P, Shah A, Tainsky M.
Assessing the Functional Bias of Commercial Microarrays
Using the Onto-Compare Database.
Biotechniques,
2003.
abstract /
pdf /
bibtex /
RIS
Draghici S, Khatri P, Bhavsar P, Shah A, Krawetz S, Tainsky M.
Onto-Tools, the Toolkit of the Modern Biologist: Onto-Express,
Onto-Compare, Onto-Design and Onto-Translate.
Nucleic
Acids Research, 2003.
abstract /
pdf /
bibtex /
RIS
Application and python library for learning Bayesian networks
from data. It's opensource and has features unmatched by
similar software. Documentation and software available at code.google.com/p/pebl-project.
anyCloud
Framework for distributed computing using a variety of clusters, grids and cloud computing platforms. Seamlessly run the same code on multiple cores and processors, an IPython cluster, an Apple XGrid and Amazon EC2 by changing a few configuration parameters. Alpha quality software available at code.google.com/p/anycloud.
A prototype web application for iteratively learning and
improving Bayesian networks. Integrated with Sakai-based portal
for authentication and data importing. Provided a novel
javascript network viewer. Screenshots coming soon.