[Product Releases]
|
|
|
|
|
[Blog]
|
|
Most
recent post
|
[News]
|
|
Can we trust docking results?
Sept 2010 IBM Systems and Technology Group releases a white paper with eHiTS and Cell
Oct 2008
EPA's ToxCastTM project will use SimBioSys' eHiTS as docking
engine
Nov, 2007
|
[Events]
|
|
243rd ACS
Mar 25-29, 2012
San Diego, CA
see >> more
|
|
|
|
|
to be held in Boston, MA, Aug 22 - 26, 2010
Presentation in TOXI
4
SESSION:General Papers (08:00 AM - 12:00 PM)
Sunday, August 22, 2010 09:00 AM
LASSO-ing potential pregnane X receptor
agonists
Dr Sean Ekins PhD, Dr Antony J
Williams PhD, Dr Zsolt Zsoldos PhD, Dr Aniko Simon PhD, Dr
Orr Ravitz PhD, Valery Tkachenko.
ChemSpider,
Royal Society of Chemistry, Wake Forest, NC, United States; ChemSpider,
Royal Society of Chemistry, Rockville, MD, United States; SimBioSys,
Inc, Toronto, ONT, Canada; Collaborations in Chemistry, Jenkintown, PA,
United States
Abstract:
The Pregnane X receptor (PXR) is a transcriptional regulator of
cytochrome P450 3A4, P-glycoprotein and many other genes involved in
metabolism and excretion. Studies have used machine learning methods to
predict PXR agonist activity for drugs but none have been made
publically available.
We utilized 203 PXR agonists (EC50 for < 10 mM)
with the Ligand Activity by Surface Similarity Order (LASSO) method, a
ligand-based tool focused on similarity of biomolecular activity rather
than structural similarity. Twenty five models were built using 8 -128
agonists. The models were tested using identification of agonists not
in the training set or using 3k, 8k and 24k drug-like decoys including
PXR inactive compounds (N=228). We found 64-128 actives provided
acceptable enrichment factors of 10% in the top 1% of compounds
screened. The best model will be deployed in ChemSpider.com, enabling
prediction of PXR agonist activity for almost 20 million compounds in
the database.
|
FullPresentation
Back to:
Fall ACS, 2010 presentations:
[1], [2], [3], [4], [5]
Back to the SimBioSys:
All Past Presentations or Upcoming Events
|
|
|
