(subtext "IBM rides again but in the wrong direction") This post provide supplementary background to three recent endeavours yours truly is currently engaged in. First is     ..read more

As some may know I have been advising  BindingDB (being on their SAB)  and the Guide to Pharmacology on medicinal chemistry patent selection and curation.  Specifically, BDB is focused on SAR extraction for the former and mapping lead compounds from the GtoPdb (often from the same patents).  The patent statistics from both sources are shown below. Note that BDB also have also curated recent WO patents and both sets of SAR are submitted to PubChem BioAssay   ..read more

 As some may know I have some personal history with BACE1  the Swiss Prot entry for which (P56817)  includes one of my highest cited publications (but sadly on the wane)  ..read more

 As we know, these days  a lot  of AI drug discovery companies are "Talking the Talk"  on social media,  However, far fewer are "Walking the Walk" in terms of tangible outputs either as full papers detailing experimental support for their particular brand of AI-magic and/or published patents with exemplified chemistry and real SAR data against declared targets.  At the bottom of this post I have pasted a series of posts related to digging out INS018_055 with the welcome engagement of GK and CG. The paper in question is "A small-molecule TNIK inhibitor targets fibr ..read more

One of the bits of good news to kick off this year was the announcement that the FDA approved no less than 55 novel therapeutics in 2023, the second highest count in the past 30 years.   Since these will be extensively covered by the Guide to Pharmacology I decided this year to just see if I could find the fist-filings with SAR data sets that could eventually be curated by BidingDB if not done by them already.  As we know the analogue SAR for approved drugs is valuable for many reasons but walking back to the first-filings to find the data can be a challenge espcially since ma ..read more

This recent paper "Deep learning-guided discovery of an antibiotic targeting Acinetobacter baumannii" has generated a lot of media interest (Guardian version below)  However I found a few quirks related to the associated PubChem entries and assay reults.  First off this image of  RS10289,   assigned the synonym by the authors of "abaucin", shown below, turns out to be a really old chestnut  from 2006 (even if AI did "find" it!) By searching RS10289  corroborated by a DECIMER run on the image there were two unequivocal  hits in PubChem, of which C ..read more

 Sine it was Rare Disease Day last week I was reminded of something I shared with the ELIXIR  Rare Diseases Community some years back that I can expand on. Very few folk were aware of this back then and it has zero social media surfacing (in my bubble at least)  so I suspect it is an under exploited resource. The link in question, FDA Search Orphan Drug Designations and Approvals  interface is shown below  plus the results for Approved Products and an example entry below that.  So far so good but so what?  Firstly it was a surprise to see over ..read more

  ..read more

 As a prelude to this Friday's (closed) Joint Meeting of NC-IUPHAR and the British Pharmacological Society for the database report plus slides (which will get post -meeting open links on the website) this is short post on SID tagging.  This is intended to be user-orientated, those interested in the technicalities are welcome to contact Simon from the database team.  Looking at the tags we have introduced will make this clearer. For starters, as many may know, as of  release 2022-3 GtoPdb has 11539 PubChem substance (SID) records.  Looking ..read more

  The good news is that database above is well descrbed in PMID35616633 bu the strcutrues are also now in PubChem.  The less good news is that these are indexed as annotations and not submitted as a selectable source per se. Notwithstanding, they can be selected via the TOC below. First thing to note is quite a significant collapse from 276,518 compounds in the interface count vs 197,124 for the distinct CID count.  Reasons not quite clear but could be chemistry rule and synching differences.   The basic PubChem stats with my custom filters on th ..read more