Time for Something New - Topomer CoMFA and Topomer Search
3D QSAR is almost twenty years old now. It has been found quite generally effective in graphically summarizing the SAR observations for a moderate number of structurally comparable ligands, and the publication record also indicates, rather consistently, that nothing is better than 3D-QSAR for forecasting potencies. In principle, ranking candidates by 3D-QSAR should be helpful in many lead optimization situations.
But 3D-QSAR’s weaknesses have persisted. Nothing useful can happen until the ligands are aligned, a tedious and subjective activity usually aimed at maximizing q2, which unfortunately is not a very robust parameter. Next, any candidates then to be ranked must also be aligned, but now necessarily without any experimental feedback at all!
I started working on topomers over twelve years ago, to support Tripos’s various screening library businesses. The modest initial goal was shape “diversity”, or minimal redundancy, in the design of these libraries. However, positive experiences soon motivated their application to “lead hopping”, by using similarity in topomer shape as a predictor of similarity in biological properties. “Lead-hopping”, to identify fragments as well as whole structures, has now been made generally available in SYBYL’s Topomer Search module.
A particularly unexpected benefit of topomers is the discovery that topomers constitute robust alignments for 3D-QSAR, as newly provided by SYBYL’s Topomer CoMFA® module. Topomer CoMFA provides results in minutes rather than weeks, and, we believe, will allow virtual screening of your entire corporate database for promising R-group ideas. Topomer CoMFA thus converts the competitive weaknesses of 3D-QSAR into competitive advantages. Isn’t it time to try something so really and so usefully new in 3D-QSAR?
I recently recorded webcasts with Brian Masek, Ph.D., Product Manager at Tripos, in which we discussed Topomer Search and Topomer CoMFA in more detail. If you haven’t viewed them yet, you can find them at:
Topomer Search: www.tripos.com/topsearchwebcast
Topomer CoMFA: www.tripos.com/topcomfawebcast
I’d like to invite you to post questions or comments related to Topomer Search, Topomer CoMFA and the webcasts here on this weblog. I’m particularly interested in hearing your thoughts and feedback on a number of questions:
- Is there value in building local models to cover not just the target biological activity, but to cover the desired profile of various biological properties? For example, models for not just activity at the target receptor, but also models to predict selectivity vs. related targets, or other off-target effects? Can a fast automated 3D QSAR tool like Topomer CoMFA help? How important is it to model and predict the level of activity vs. a simple active/inactive prediction?
- The automated alignments generated by topomer technologies provide a number of benefits. Speed, automation, and ease of use. Because alignment is automated, Topomer CoMFA models can be used for virtual screening (coming soon) to suggest new R-groups or substituents that will improve biological properties; which is very valuable for lead optimization. However, the automated alignments generated by topomers, while useful for building predictive models, are not necessarily the physical or biological alignments. I believe the benefits to discovery projects outweigh these concerns. Your thoughts?
We’ll be monitoring this blog and I’ll respond periodically with answers and comments. I hope this weblog will be a useful public forum for discussion and feedback.
- Dick Cramer
Leave Your Comments ...
You must be logged in to post a comment.