Sign in

Count F.P. Morgan

@ecfp.bsky.social
192 followers 691 following 52 posts

cheminformatics @ všcht prague molecular.beauty github.com/dehaenw

PostsRepliesMedia
Count F.P. Morgan @ecfp.bsky.social · 24/08/2026
here's a mirror of what i could find of the tenderbutton blog on archive.org now made fast and searchable in one place including comments (some missing images in newer posts) molecular.beauty/tenderbutton... enjoy! #chemsky #chemchat
molecular.beauty
Tenderbutton — a reading archive
A readable preservation of the Tenderbutton chemistry blog and its comments.
011
Count F.P. Morgan @ecfp.bsky.social · 20/08/2026
does any have the tenderbutton/dylan stiles archive in one place? #chemsky #chemchat There's a lot mirrored on archive.org but it would be great to have it fast, searchable and ideally complete somewhere
201
Count F.P. Morgan @ecfp.bsky.social · 30/07/2026
now with fixed stereochemistry, easy installation via pip and large validation (99.998% roundtrip success on chembl) github.com/dehaenw/blincs
github.com
GitHub - dehaenw/blincs: Breadth-first Line Notation for Chemical Structures​
Breadth-first Line Notation for Chemical Structures​ - dehaenw/blincs
000
Count F.P. Morgan @ecfp.bsky.social · 26/05/2026
the "heck" reaction in the abstract image doesn't make any sense for that isomer of benzazepine. Fig 3.2.1. also has a nonsensical molecule. Both are obvious GenAI. Authors, reviewers and editors should have done a bit better effort to at least factcheck and clean up the mistakes in the GenAI output
001
Count F.P. Morgan @ecfp.bsky.social · 30/03/2026
new preprint, enjoy filtering and generating molecules! chemrxiv.org/doi/full/10....
chemrxiv.org
LACAN: Leveraging adjacent co-occurrence of atomic neighborhoods for molecular scoring and generation Authors | ChemRxiv
LACAN (Leveraging Adjacent Co-occurrence of Atomic Neighborhoods) is a bond-level statistical scoring function and filter based on adjacent atomic environment co-occurrence. When integrated into evolutionary search it enables chemically constrained ...
000
Count F.P. Morgan @ecfp.bsky.social · 09/02/2026
sad to see the low success rates of MSCA this year, with many excellently rated projects not meeting the unusually high funding threshold. wonder if the big increase in applications is due to the decline of american science funding and prestige
020
Reposted by Count F.P. Morgan
Manas Mahale @manasmahale.bsky.social · 31/12/2025
literally no one asked, but I started a newsletter. it’s called "This Week In Cheminformatics" (very original, I know). First issue is out now. have a read and a very happy new year ! thisweekincheminformatics.substack.com/p/one #cheminformatics
031
Count F.P. Morgan @ecfp.bsky.social · 29/12/2025
nitrogen missing on the caff molecule unfortunately
120
Count F.P. Morgan @ecfp.bsky.social · 18/12/2025
using 1999bit fingerprints so fits in the <2000 ft tabpfn limit. feels dirty
000
Count F.P. Morgan @ecfp.bsky.social · 13/12/2025
looks great, though they definitely broke the scifinder terms with this, at least as of two years ago some scifinder people who visited our uni told me "curated dataset creation for cheminformatics" is forbidden (i asked)
000
Count F.P. Morgan @ecfp.bsky.social · 17/11/2025
need this custom hat i found on the Ken Houk group website #chemsky
A black hat with an illustration of the transition state of the 6+4 cycloaddition between tropone and dimethylfulvene, and a black beanie that reads "55 years of the [6 + 4] reaction"
0160
Count F.P. Morgan @ecfp.bsky.social · 23/10/2025
pubs.acs.org/doi/10.1021/...
010
Count F.P. Morgan @ecfp.bsky.social · 07/10/2025
CADD milestone for me: glad to have our first compounds in the PDB (virtual screened by our team during CACHE#3) : 7HQH, 7HQM, 7HQL, 7HQE, 7HPL, 7HPS, 7HQ5, 7HQO, 7HQP, 7HPY, 7HPX, 7HQ8
000
Count F.P. Morgan @ecfp.bsky.social · 04/10/2025
haven't tried it myself, but saw a presentation on this one by @talagayev.bsky.social and it looked good github.com/wolberlab/Op...
github.com
GitHub - wolberlab/OpenMMDL: Interface to OpenMM for easy setup of molecular dynamic simulations of protein-ligand complexes
Interface to OpenMM for easy setup of molecular dynamic simulations of protein-ligand complexes - wolberlab/OpenMMDL
010
Count F.P. Morgan @ecfp.bsky.social · 04/10/2025
openmm is great for this purpose, good python bindings and easy installation (for md standards...)
100
Count F.P. Morgan @ecfp.bsky.social · 19/09/2025
that's not the SMILES of oxalic acid, and the structure shown also isn't oxalic acid nor carbonate.
100
Count F.P. Morgan @ecfp.bsky.social · 14/09/2025
sometimes, you simply require thousands of unique SMILES for the same molecule (e.g. teaching a language model to pretend it's a graph based model instead). A prototype of the trivial "solution" to this problem using disconnected atoms dialect of SMILES is here now: github.com/dehaenw/ultr...
github.com
GitHub - dehaenw/ultra_enumerate: enumerate a large amount of rdkit valid SMILES.
enumerate a large amount of rdkit valid SMILES. Contribute to dehaenw/ultra_enumerate development by creating an account on GitHub.
020
Count F.P. Morgan @ecfp.bsky.social · 08/09/2025
Paraquat might be unstable upon pyrolysis. Turns to less toxic bipyridine apparently. Upon looking it up, toxicity to dopaminergic neurons of paraquat has been extensively studied and was shown to be real, so while I can't speak for the real PD risk, the biological proof-of-concept is there
000
Count F.P. Morgan @ecfp.bsky.social · 08/09/2025
it seems obvious it could cause PD, given its similarity to MPP+
100
Count F.P. Morgan @ecfp.bsky.social · 20/08/2025
yes, me. i used marvinsketch for some time because it was linux friendly
010
Count F.P. Morgan @ecfp.bsky.social · 06/08/2025
github.com/HaolinZhan/P... (agree that it's weird it's not linked by the journal, and not claiming a few example matlab files are satisfactory)
github.com
GitHub - HaolinZhan/PE-NMR
Contribute to HaolinZhan/PE-NMR development by creating an account on GitHub.
000
Reposted by Count F.P. Morgan
Greg Landrum @greglandrum.bsky.social · 17/07/2025
Today's #RDKit blog post is a heartfelt plea for clearer communication. greglandrum.github.io/rdkit-blog/p...
greglandrum.github.io
Please stop saying “The Tanimoto similarity is” – RDKit blog
A simple tip to explain what you actually did
2316
Count F.P. Morgan @ecfp.bsky.social · 22/06/2025
out now in JCIM pubs.acs.org/doi/10.1021/...
pubs.acs.org
CACHE Challenge #2: Targeting the RNA Site of the SARS-CoV-2 Helicase Nsp13
AbstractA critical assessment of computational hit-finding experiments (CACHE) challenge was conducted to predict ligands for the SARS-CoV-2 Nsp13 helicase RNA binding site, a highly conserved COVID-1...
000
Count F.P. Morgan @ecfp.bsky.social · 12/05/2025
A rare pic of Taffee Tanimoto (of Tanimoto similarity fame) in front of the classroom ca 1966. Source: web.archive.org/web/20040906...
Professor Taffee Tanimoto in class with his students at the former Park Square campus, circa 1966
000
Count F.P. Morgan @ecfp.bsky.social · 02/05/2025
One for the #Alphafoldogy lovers: Reproducible severe ligand hallucination that commonly occurs in AF3-likes: cyclohexane to benzene Boltz (1.0.0) input >A|smiles C1CCCCC1CCN(C1CCCCC1)CC1CCCCC1 output in the pic below
010
Count F.P. Morgan @ecfp.bsky.social · 25/04/2025
just sneakily registered for the #ICCS2025 , looking forward to impressing everyone except dutch people by impeccably pronouncing noordwijkerhout in verkavelingsvlaams
000
Count F.P. Morgan @ecfp.bsky.social · 10/04/2025
it's a "cautionary principle" model - assumes every unseen chemical is toxic just to be sure!
110
Count F.P. Morgan @ecfp.bsky.social · 18/03/2025
RDKit UGM 2025 in Prague (at VŠCHT) !
000
Count F.P. Morgan @ecfp.bsky.social · 15/03/2025
Baseline approaches in the @polarishub.io x ASAP discovery x OpenADMET challenge (in advance of next week's leaderboard...) molecular.beauty/blog/2025/03...
molecular.beauty
”Baseline approaches in the Polaris challenge”
Molecular property prediction has a serious bottleneck: data. Commonly used benchmark data have various problems as talked about at length elsewhere. One organization which is trying to tackle this pr...
010
Count F.P. Morgan @ecfp.bsky.social · 13/03/2025
nice, as the product is a michael acceptor too it might even be possible to conjugate with a second AA...
000
Count F.P. Morgan @ecfp.bsky.social · 10/03/2025
Hyperzrone A: great structure and great name to boot
020
Count F.P. Morgan @ecfp.bsky.social · 07/03/2025
glad to be on the CACHE #2 preprint chemrxiv.org/engage/chemr...
chemrxiv.org
CACHE Challenge #2: Targeting the RNA Site of the SARS-CoV-2 Helicase Nsp13
A critical assessment of computational hit finding experiments (CACHE) challenge was conducted to predict ligands for the SARS-CoV-2 Nsp13 helicase RNA binding site, a highly conserved COVID-19 target...
010
Count F.P. Morgan @ecfp.bsky.social · 07/03/2025
The activity of this new compound is unknown, but Taxol itself has low oral bioavailability and is actually not that potent, being dosed at 100-175 mg/m2 IV. Seems like it wouldn't be that difficult to safely deal with in an academic lab.
000
Count F.P. Morgan @ecfp.bsky.social · 02/03/2025
Chemical space(s) molecular.beauty/blog/2025/03...
molecular.beauty
Chemical space(s)
An interesting perspective on “Chemical Space” was penned by Jean-Louis Reymond recently, titled “Chemical space as a unifying theme for chemistry”. Reymond argues for a total, unifying “Chemical Spac...
030
Count F.P. Morgan @ecfp.bsky.social · 22/02/2025
The "FedEx arrow" of cheminformatics is the arrow in the letter D in the RDKit logo...
000
Count F.P. Morgan @ecfp.bsky.social · 21/02/2025
TGIBF₃
032
Count F.P. Morgan @ecfp.bsky.social · 09/02/2025
submitting something tonight!
000
Reposted by Count F.P. Morgan
Karel "CaCO3" Berka @krapnik.bsky.social · 26/01/2025
Tomorrow starts the 8th Advanced in Silico Drug Design Workshop #8ADD www.kfc.upol.cz/8add organized by Pavel Polishchuk from IMTM LF UP, me from @fchupol.bsky.social, our colleagues, and students and supported by @elixircz.bsky.social .
kfc.upol.cz
8th Advanced in silico Drug Design workshop 2025 – Katedra fyzikální chemie
172
Count F.P. Morgan @ecfp.bsky.social · 23/01/2025
theres an interesting parallel on a quite fundamental level too, in the microsound area the gabor wavelet made the wave particle duality and time-freq uncertainty principle directly relevant for sound synthesis
010
Count F.P. Morgan @ecfp.bsky.social · 18/01/2025
see the original claim here, it's a bit more clear particularly when shown superimposed on the 3d structure x.com/btnaughton/s...
x.com
x.com
020
Count F.P. Morgan @ecfp.bsky.social · 10/01/2025
also rather interesting axial chirality..
110
Count F.P. Morgan @ecfp.bsky.social · 07/01/2025
is there currently an active leaderboard (guacamol style) for molecular generators?
000
Count F.P. Morgan @ecfp.bsky.social · 29/12/2024
that footage looks strange to me. a 10 kg fentanyl operation wouldn't use open fire heating of completely full cooking pots of some slurry. maybe this is rather some small shrimp formulation operation or pranksters making bechamel
110
Count F.P. Morgan @ecfp.bsky.social · 27/12/2024
double bond in the middle of the fused heterocycle is missing
100
Count F.P. Morgan @ecfp.bsky.social · 22/12/2024
impressive amount of distinct Li's!
010
Count F.P. Morgan @ecfp.bsky.social · 13/12/2024
ironic that the thing that might save us from from the mirror germ apocalypse is gonna be an L-ala-L-ala mimic beta lactam
000
Count F.P. Morgan @ecfp.bsky.social · 12/12/2024
RIP to a giant chem-bla-ics.linkedchemistry.info/2024/12/08/r...
chem-bla-ics.linkedchemistry.info
Richard L. Apodaca
If you are into openscience chemistry or chemistry blogging, then you probably heard of Rich Apodaca’s Depth-First blog. Rich started blogging in 2006 but this is not how I discovered his work origina...
000
Count F.P. Morgan @ecfp.bsky.social · 10/12/2024
all the hits (including ours 😁) except one are based on one of two pre-existing scaffolds from the reference PNAS paper (and decorated sparingly), so poses should be rather good just from the starting point of aligning to the template
000
Count F.P. Morgan @ecfp.bsky.social · 10/12/2024
we (#1715) placed 5th in the #CACHE3 challenge! (a lot of hits, but penalized due to novelty) As a cheminformatician, I love to see successful near neighbor behavior. Congrats to all involved and congrats to the 7-deazapurine and pyrimido[4,5b]indole scaffold. cache-challenge.org/results-cach...
cache-challenge.org
Results of Cache Challenge # 3 | CACHE
Accelerating the impact of AI-driven Drug Discovery
000
Count F.P. Morgan @ecfp.bsky.social · 07/12/2024
imo not a ideas vs solutions issue but a arrogant ML researchers vs. the rest issue. they could've carefully reported their results, they didn't. i know of several people that wasted a substantial amount of time confirming diffdock is useless for any CADD task. good this paper is here to warn others
020