Chris de Graaf

Identification of nanomolar adenosine A2A receptor ligands using reinforcement learning and structure-based drug design - Nature Communications Here the authors combine a deep generative model with structure-based drug design and prospectively validate functionally active, nanomolar, A2A adenosine receptor ligands and solve their crystal stru...

Closing the loop on #GenAI for #GPCR #SBDD

www.nature.com/articles/s41...

tinyurl.com/yeyhfr7j

Identification of nanomolar adenosine A2A receptor ligands using reinforcement learning and structure-based drug design - Nature Communications Here the authors combine a deep generative model with structure-based drug design and prospectively validate functionally active, nanomolar, A2A adenosine receptor ligands and solve their crystal stru...

Closing the loop on #GenAI for #GPCR #SBDD

www.nature.com/articles/s41...

tinyurl.com/yeyhfr7j

Stand Up And Be Counted

Biopharma companies and CEOs are keeping their heads down at their own peril. They should speak up about what’s happening to the NIH and other science agencies before it’s too late.

Silence gives consent. And no one should consent to this.

Protein structure embeddings reveal undersampled and de novo structure space. (B) First two principal components of mean-pooled ESM3 embeddings colored by helix content determined by DSSP. The indicated dashed guide lines denote visual boundaries of native structure space not sampled (Undersampled) and novel regions of protein structure for this space only observed in samples but not in native structures (De novo). (C) Rasterized visualization of panel B with 16 equally spaced grid squares in each principal component axis. A representative structure from each grid was chosen at random. Empty grid squares indicate the absence of any structure in the enclosed region. De novo alpha helices are shaded along the lower-right diagonal and the structures from CATH which do not have corresponding structures in the samples are shaded along the left and top rims. The structures are displayed in CATH raster plot are given in the Supplementary Information.

Protein backbone diffusion models consistently undersample catalytically important motifs, and oversample idealized helices, raising questions about the appropriateness of these methods in designing starting points for enzyme design & evolution. From www.biorxiv.org/content/10.1...

If you are interested in #AI and #DrugDiscovery, attend the upcoming ICCS June 1st - Jun 5th in Noordwijkerhout!

Last call for papers. Please repost!

Registration and abstracts (deadline 14 Feb): iccs-nl.org/general-info...

Many thanks @gati.bsky.social!

Structural and functional determination of peptide versus small molecule ligand binding at the apelin receptor - Nature Communications This study explores apelin receptor’s role in cardiovascular function, identifying residues critical for binding through genetic variants, AlphaFold2 modelling and base editing in cardiomyocytes. Co-c...

#StructuralBiology insights into #Genetic #GPCR variations identified in the UK 100,000 Genomes Project with differentiating effects on peptide (affected) vs. small molecule agonists (not affected) #pharmacology

www.nature.com/articles/s41...

@baoilleach.bsky.social on #Bluesky!

@baoilleach.bsky.social on #Bluesky!

This is figure 1, which shows the geographical distribution of modern human specimens older than 40 ky that produced genome wide data

Analysis of the oldest-known genomes from early modern humans in Europe, who lived around 45,000 years ago, helps to provide a more precise date for when Neanderthals and modern humans mixed, according to research published in Nature. https://go.nature.com/3ZR8Kh1 🧪 🏺

Here's a nice Christmas gift - ChEMBL 35 is out! Use your well-deserved Christmas holidays to spend time with your loved ones and explore the new release of ChEMBL 35!            This fresh...

ChEMBL 35 is out. Happy Holidays!
chembl.blogspot.com/2024/12/here...

Reply by Jain and colleagues on the response of Corso and colleagues:.
More response to the response (DiffDock/DiffDock-L)
www.linkedin.com/pulse/more-r...

Chimera: Accurate retrosynthesis prediction by ensembling models with diverse inductive biases Planning and conducting chemical syntheses remains a major bottleneck in the discovery of functional small molecules, and prevents fully leveraging generative AI for molecular inverse design. While ea...

new preprint on chemical synthesis ML models

- showing how to combine multiple models in a principled way
- modern Transformers + GNN to featurize chemical reaction:
- new insights in where the models shine
+ bonus: find the quirky named reaction!

Feedback welcome!

arxiv.org/abs/2412.05269

We keep getting messages saying climate change is a hoax and the world is flat. Is it 2024?!

Of: Alles voor zoete koek slikken

Moving #CADD #AI for #GPCR #SBDD posts and discussions to #Bluesky !

bsky.app/profile/cdg-...

Comparative Study of Allosteric GPCR Binding Sites and Their Ligandability Potential The steadily growing number of experimental G-protein-coupled receptor (GPCR) structures has revealed diverse locations of allosteric modulation, and yet few drugs target them. This gap highlights the...

Comparative Ligandability Analysis and new Nomenclature Allosteric GPCR binding sites

pubs.acs.org/doi/10.1021/...

tinyurl.com/ybthux9c

#GPCR #CADD #SBDD #ALLODD

Modern hit-finding with structure-guided de novo design: identification of novel nanomolar adenosine A2A receptor ligands using reinforcement learning Generative chemical language models have demonstrated success in learning language-based molecular representations for de novo drug design. Here, we integrate structure-based drug design (SBDD) princi...

Closing the CADD AI SBDD loop - From GPCR structures to RNN de novo drug design and back again!

chemrxiv.org/engage/chemr...

#GPCR #AI #RNN #CADD #SBDD

The first paper from the Small Molecule Steering Committee at Polaris introduced practical guidelines for method comparison in ML-driven drug discovery.

We'll be discussing this live on Dec 5th from 11 - 12 PM ET.

Join the convo: portal.valencelabs.com/events/post/...

Bijvoorbeeld: Koekje van eigen deeg

🙋‍♂️👨‍💻

#CASP16 results are online at the prediction center website - conference starting tomorrow evening.

Have a safe trip - see you soon!

predictioncenter.org/casp16/

More generally (not for HIV protease case specifically) I would propose to consider 4-5 options:

1) Displace
Energetically unfavorable water in lipophilic region with lipophilic ligand moiety

2) Stabilise
Design favorable ligand-water-protein mediated interaction by introduction polar ligand moiety

3) Replace
Polar ligand moiety mimicking energetically favorable water molecule

4) Leave alone
a) Energetically favorable molecule
b) Bulk water/accessible region / no need to expand ligand in this direction unless for eg solubilising groups

4-5 options/scenarios/sttategies how to deal with water in #SBDD #CADD #MedChem ?

@darrylbmcconnell.bsky.social