@bioconductor 3.22 is out, including BridgeDbR 2.22: https://bioconductor.org/packages/release/bioc/html/BridgeDbR.html

This release improves handling of ChEBI compact identifiers from @bioregistry, improves the vignette, and adds a second vignette, and explains more about rJava installation.

See https://genomic.social/@bioconductor/115467708443587417 for the general announcement

RE: https://fosstodon.org/@gicrisf/115515504598989643

that sounds useful! are there other #cheminformatics toolkits integrated into #Emacs ?

After several months, I'm finally sharing the first pre-release of Emacs Indigo, a native #Emacs module that brings the #Indigo #cheminformatics library to Emacs #Lisp.

It's now at a point where the core functionality is solid enough to share. (1/7)

BChemXtract, the new @BeilsteinInstitut ChemDraw parser and structure extractor, uses CDK 2.11

<properties>
<cdk.version>2.11</cdk.version>
</properties>

https://hessen.social/@BeilsteinInstitut/115497448450303612

BChemXtract - a ChemDraw parser and structure extractor - is our first open source software project and is available on @github:

https://github.com/Beilstein-Institut/BChemXtract

Reflecting on a talk from earlier this year: “hard-to-make” molecules still trip up today’s tools. Useful reminder to pair algorithms with chemist intuition and better datasets. Anyone else see this in practice?

#retrosynthesis #computationalchemistry #cheminformatics #AI

https://idees.moleculair.es/retrosynthesis

interesting to see that also Journal of Cheminformatics (still not on Bluesky or Mastodon?) articles are also showing up in podcasts :) The most recent related to IP concerns: https://doi.org/10.1186/s13321-025-00982-w

HT @altmetric.com

Our new paper, MARCUS (Molecular Annotation and Recognition for Curating Unravelled Structures), has been published in the Digital Discovery Journal.

Paper: https://doi.org/10.1039/D5DD00313J

We built MARCUS—a free, open-source AI platform that extracts millions of chemical structures trapped in PDFs in minutes, rather than hours.

I heard that Antony Williams won the 2026 Skolnik Award! https://www.linkedin.com/posts/drbobclark_dr-antony-williams-herman-skolnik-awardee-activity-7381513979346653184-YSkq

Congrats, Tony! (is he not on Mastodon?)

Does anyone here have access to "Substructure Retrieval by Means of the BASIC Fragment Search Dictionary Based on the Chemical Abstracts Service Chemical Registry III System", JCICS (1979), https://pubs.acs.org/doi/pdf/10.1021/ci60017a013 ?

I want to know how many fragments it used.

I know it had 1942 fragment numbers, but multiple fragments were mapped to the same fragment number. (Eg, the 3rd BASIC dictionary had over 5,000 fragments and 2047 fragment numbers.)

#cheminformatics @egonw ?

The evaluation&implementation of any methodology must be based on how well it provides the chemist with the best information for the least cost. This can only be done if the problem is approached, not only as a chemical problem, but also as a computer problem since the computer is the medium through which this information must be processed. Chemical data representation&its subsequent processing must always be considered in terms of computer adaptability&efficiency. —Penny, 1965 #cheminformatics

anyone with a #hacktoberfest repo in the area of #chemistry, #biology, #cheminformatics, or #bioinformatics where I can make a PR?

Identification of FtsZ Protein Nucleotide-Binding Site Effectors Based on Cheminformatics and Structural-Biological Analysis - #FtsZ #nucleotidebindingsite #interdomaincleft #effectors #ligand–proteininteraction #FragFp #pharmacophoresearch #moleculardocking #artificialintelligence #cheminformatics #structuralbiologicalanalysis - https://link.springer.com/article/10.3103/S0095452725040073

Any ideas how to better promote #chemfp? I'm only on the Fediverse with only a handful of #cheminformatics posters (including a couple of chemfp users!). Mailing lists are almost dead.

At ICCS many knew of me, but I found no leads.

There's all sort of great stuff in chemfp, but as they say, great marketing beats great engineering ... and evidence shows I'm not a good marketer.

I've tried cold emailing relevant people. That didn't work.

Ideas? Mail physical fliers? Skywriting over Cambridge?

I've just released chemfp 5.0, the latest version of my #cheminformatics fingerprint package, with command-line tools and a full #python API.

The chemfp home page is https://chemfp.com/

The major new features are: 1) FPB file support 1B+ fingerprints, 2) sharded similarity search, 3) initial support for count fingerprints, and 4) OEChem and RDKit implementation of the Klekota-Roth fingerprints.

Details of the new features are at https://chemfp.com/docs/whats_new_in_50.html .

No cost for academic use!

Our new paper now Published: Cheminformatics Microservice V3!

Paper: https://doi.org/10.1186/s13321-025-01094-1

Web app + API: RDKit, CDK, Open Babel — NO install, 100% open-source.

Try now: app.naturalproducts.net

ChEMBL 36 is out! if you're using chembl-downloader for all of your ChEMBL needs, then you just have to re-run your reproducible workflows and get arbitrarily new and better results, you beautiful nerd

https://chembl.blogspot.com/2025/09/chembl-36-is-out.html

currently preparing slides to present at the sixth @NFDI4Chem consortium meeting on using the @bioregistry for finding ontologies, standardizing cross-references, and supporting RDM

also thinking about how to automate generating slides that summarize the state of the database for future talks and reminded of https://xkcd.com/1205

Any idea how to get the (#cheminformatics) structures for a #ChEBI ontology term? Like, how do I get all the terpenoids, CHEBI:26873?

https://www.ebi.ac.uk/ols4/ontologies/chebi/classes/http%253A%252F%252Fpurl.obolibrary.org%252Fobo%252FCHEBI_26873?lang=en says there are 11,119 terpenoids but I see no way to download them.

Trying the web API, https://www.ebi.ac.uk/chebi/backend/api/public/ontology/all_children_in_path/?relation=has_functional_parent&entity=CHEBI%3A26873&three_star_only=false&page=1&size=15&download=false says there are 547 (482 if three_star_only=true).

With 'download=true' I get the data in full (unpaged) data in tsv format, with the same counts.

What am I missing?