Transcript Zhijun Li

PyMine – A PyMol Plugin to Integrate and
Visualize Chemical and Biological Data for
Drug Discovery
Zhijun Li, Ph.D.
Departments of Chemistry & Biochemistry
University of the Sciences in Philadelphia
USA
Drug discovery is a costly process
http://www.chemistry-blog.com/2012/01/04/tedtalk-medicine-for-the-99-hes-about-99-wrong/.
Cheminformatics and bioinformatics can
facilitate drug discovery

Target identification

Target modeling

In silico screening and
design
Krasky et al., Genomics, 2007, 89: 36-43.
Data integration remains a challenge
www.ncbi.nlm.nih.gov
Current Approaches for Bioinformatics Data
Integration

Service-oriented web approach

Data warehousing approach

Other approaches
PyMine Development
PyMol
http://ww.pymol.org

Molecular visualization
system

Open source

Leading software package
Databases

UniProt

IBIS

PDB

HUMSAVAR

ChEMBL

KEGG
Integration
Interface and Output
Panel Interface
PyMol 3D visualization
Chaudhari and Li., BMC Research Notes, 2015, submitted.
Output Directory
Case Study
Human Malic Acid Enzymes

Major enzymes in the production of NADPH

Also closely associated with the TCA cycle, a central
metabolic hub

In mammalian cells, two major malic enzyme
isoforms (ME1 and ME2) have been identified
Malic enzymes reciprocally module p53
Jiang et al., Nature, 2013, 493: 689-695.
PyMine Study and CADD Approach

ME1 structure

Ligand-based approach
using ShapeSignature

Catalytic site

Structure-based approach
using Glide

Known inhibitors

Assay testing
CADD Results
Summary

Developed an innovative and automated tool to integrate
and visualize chemical and biological data for drug
discovery.

PyMine integrates data from six high-quality chemical
and biological databases and maps imported structural
information onto receptor 3D structure.

PyMine imports and/or generates files that can be used
directly for drug discovery projects.
https://github.com/rrchaudhari/PyMine
Future Development

To include support for additional chemical/biological
databases.

To apply inductive logic programming in order to
generate and prioritize data driven hypotheses.

To automate cheminformatics workflows based on
imported structural data to test generated hypothesis.
Acknowledgement
Univ. of Pennsylvinia
Dr. Wenchao Song; Dr. Xiaolu Yang
CHOP
Dr. Hakon Hakonarson,
Duke University
Dr. Chin-Ho Chen; Dr. Li Huang
Wistar Institute
Dr. Jose Conejo-Garcia
University of Sydney, Australia
Dr. John Christodoulou
Meharry Medical College
Dr. Hua Xie
Univ. of Missouri
Dr. Gerald L. Hazelbauer
Usciences
Dr. Zhiyu Li; Dr. Preston Moore
Dr. Jun Gao
Dr. Vagmita Pabuwal
Dr. Jhenny Galan
Dr. Andrew Heim
Usha K. Muppirala
Rajan Chaudhari
Daoyuan Dong
Tejashree Redij
David Cho
PhRMA Foundation
NSF
NIH
UPenn- ITMAT