tag:blogger.com,1999:blog-1136153439451224584.post384920825898682522..comments2024-03-27T06:45:59.174-07:00Comments on Practical Fragments: Plenty of room at the bottom (of chemical space)Dr. Teddy Zhttp://www.blogger.com/profile/07288045760981372367noreply@blogger.comBlogger2125tag:blogger.com,1999:blog-1136153439451224584.post-53163852766704737652014-11-26T14:01:41.384-08:002014-11-26T14:01:41.384-08:00I really enjoyed the paper, but I don't quite ...I really enjoyed the paper, but I don't quite agree some of with the authors points with respect to the 3D fragments. The authors state that even for PPI targets, there is still a higher % of planar X-ray hits than the % in the library itself (although . . . not statistically significant)". Then the authors state "non-planar fragments . . . generally show a slightly poorer hit rate than flat molecules".<br /><br />My interpretation of the data shown is that for kinases, the aromatic hit rate is higher. But for PPI's, the hit rates of planar and non-planar hits appear pretty comparable.<br /><br />I really like the point about trying to de-emphasize hit rates from fragment libraries and instead to focus on whether or not the hits "can be optimised into good chemical leads or drugs".<br /><br />Along those lines, for very difficult targets with shallow 3D pockets, the hit rates of non-planar molecules ought to be quite low, but perhaps with <i>higher</i> hit-rates than aromatics, as the complementarity of aromatics may be too weak to detect binding. However, this would depend on the fragment library containing the right non-planar fragments that can fit...a tough challenge given the much larger chemical space of non-planar fragments.<br /><br />Thus, hit rates can be quite deceiving for that reason as well. If you happen to have the right fit for a tough target with a shallow, spherical pocket, hit rates might be exceedingly low, but also of tremendous value and able to generate leads. And in those cases, aromatics might not show detectable binding. Ind if your frag-lib's non-planar space is insufficient, none of the non-planars might be detectable as well, but if you've got the right substituted, saturated ring, you'll get the hits.<br /><br />Lastly, I really like the discussion of how Astex found the chemical space coverage of commercially available fragments to be sparse, thus propelling them to "enhance the library via synthesis".Matthew R. Leehttps://www.blogger.com/profile/09787307466209873125noreply@blogger.comtag:blogger.com,1999:blog-1136153439451224584.post-87177536464360517262014-11-10T13:38:34.863-08:002014-11-10T13:38:34.863-08:00This paper contains one of the least scientificall...This paper contains one of the least scientifically relevant passages I have read in a long time. <br /><br />"In order to assess the deviation from planarity for molecules we use an approach that has recently been described and analysed in detail by Firth et al. (Firth et al., 2012). A number of years ago, we independently developed this method and here we give a brief description of the approach."<br /><br />I think I'll try to claim gravity the next time I'm writing up a manuscript!<br />Anonymousnoreply@blogger.com