Congrats Sukrit! Excited to read!

G.L.O.S.S. - We Live YouTube video by angel

Trans day of G.L.O.S.S rips!

www.youtube.com/watch?v=v2wa...

Big week!

Pleasantly surprised this morning to wake up to both the first preprint of my postdoc and first preprint using my new name :)

We used a combination of experiment and simulation to tease out the mechanistic basis of specific recognition of FcγRIIa.

www.biorxiv.org/content/10.6...

I'd like to propose that we stop inventing new ways to log in

Has anyone else noticed that the bathrooms at #bps2026 have low-key good ambient electronic music for no reason?

I'm giving a poster later today @ 1:45. Come learn about how we're using MD and free energy calculations to better understand specific antibody-epitope recognition in concert w/ experiment! #bps2026 #BPS2026

@filizolalab1 kicking off at #bps2026 with a poster presentation by Kirill, on a collaborative work with the Skiniotis and Robertson labs. Go listen to Kirill!

BOLD-GPCRs: A Transformer-Powered App for Predicting Ligand Bioactivity and Mutational Effects across Class A GPCRs G Protein-Coupled Receptors (GPCRs) are important targets for drug discovery owing to their ability to respond to a broad range of stimuli and their involvement in numerous pathologies. Although traditional ligand-based and structure-based approaches have facilitated the development of effective therapeutics for many GPCRs, these approaches often fall short when applied to receptors with limited ligand or structural data. This limitation highlights the critical need for advanced strategies capable of accurately predicting ligand bioactivity across the entire GPCR family, especially for understudied receptor subtypes. In this study, we introduce BOLD-GPCRs (BERT-Optimized Ligand Discovery for GPCRs), a deep learning framework designed to enhance the prediction of ligand bioactivity across class A GPCRs. Accessible via a user-friendly web interface, BOLD-GPCRs employs transfer learning and leverages curated data sets of known class A GPCR ligands, receptor sequences, and signaling-relevant mutations. By integrating dense neural network classifiers with transformer-based protein language models, BOLD-GPCRs captures complex relationships between receptor sequence/function and ligand activity. Our results demonstrate that BOLD-GPCRs achieves robust predictive performance for both ligand bioactivity and mutational effects across a broad range of class A GPCRs, underscoring its potential as a valuable tool for ligand discovery, especially for poorly characterized receptors.

After the most surreal peer review process we are out in the wild with BOLD-GPCRs: A Transformer-Powered App for Predicting Ligand Bioactivity and Mutational Effects across Class A GPCRs | Journal of Chemical Information and Modeling pubs.acs.org/doi/full/10....

First preprint of the year, led by Junjie Zhu from Haifeng Chen’s lab

Extending Conformational Ensemble Prediction to Multidomain Proteins and Protein Complex

doi.org/10.64898/202...

Large-scale collaborative assessment of binding free energy calculations for drug discovery using OpenFE Accurately measuring compound binding affinities is key to driving the pharmaceutical development process. Rigorous physics-based in silico approaches, particularly alchemical free energy methods, hav...

OpenFE is ready for production! chemrxiv.org/engage/chemr...

In collaboration with our industry partners, we ran benchmarking simulations of our hybrid-topology RBFE protocol on a large collection of both public and private protein-ligand binding datasets.

#compchem

"Academics/scientists, stop using AI generated images" challenge, any%

Desireless performing in 1986

Physical Chemist at the Biochem conference #chemchat

It’s official: ChatGPT can't draw a GPCR, but you’ll master them in the Filizola Lab
😜 Join Us! (send your CV and the names of at least two references to my institutional email) #Postdoc #scientist #research

Don’t let generative AI shape how we see chemistry | Nature Reviews Chemistry When generative artificial intelligence (AI) draws molecules, it doesn’t just get the science wrong — it risks reshaping how we see chemistry, spreading incorrect representations of our world that mislead learners, distort peer-reviewed reports, and erode trust in scientific work and rigour.

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

How many crystal structures do you need to trust your docking results? www.biorxiv.org/content/10.1101/2025.09....

Our first protein design paper out in Protein Science
onlinelibrary.wiley.com/doi/10.1002/...

This seems super cool, excited to give it a read!

Awesome! I'm excited to give these a read and try this out!

Abstract: Under the banner of progress, products have been uncritically adopted or even imposed on users — in past centuries with tobacco and combustion engines, and in the 21st with social media. For these collective blunders, we now regret our involvement or apathy as scientists, and society struggles to put the genie back in the bottle. Currently, we are similarly entangled with artificial intelligence (AI) technology. For example, software updates are rolled out seamlessly and non-consensually, Microsoft Office is bundled with chatbots, and we, our students, and our employers have had no say, as it is not considered a valid position to reject AI technologies in our teaching and research. This is why in June 2025, we co-authored an Open Letter calling on our employers to reverse and rethink their stance on uncritically adopting AI technologies. In this position piece, we expound on why universities must take their role seriously toa) counter the technology industry’s marketing, hype, and harm; and to b) safeguard higher education, critical thinking, expertise, academic freedom, and scientific integrity. We include pointers to relevant work to further inform our colleagues.

Figure 1. A cartoon set theoretic view on various terms (see Table 1) used when discussing the superset AI (black outline, hatched background): LLMs are in orange; ANNs are in magenta; generative models are in blue; and finally, chatbots are in green. Where these intersect, the colours reflect that, e.g. generative adversarial network (GAN) and Boltzmann machine (BM) models are in the purple subset because they are both generative and ANNs. In the case of proprietary closed source models, e.g. OpenAI’s ChatGPT and Apple’s Siri, we cannot verify their implementation and so academics can only make educated guesses (cf. Dingemanse 2025). Undefined terms used above: BERT (Devlin et al. 2019); AlexNet (Krizhevsky et al. 2017); A.L.I.C.E. (Wallace 2009); ELIZA (Weizenbaum 1966); Jabberwacky (Twist 2003); linear discriminant analysis (LDA); quadratic discriminant analysis (QDA).

Table 1. Below some of the typical terminological disarray is untangled. Importantly, none of these terms are orthogonal nor do they exclusively pick out the types of products we may wish to critique or proscribe.

Protecting the Ecosystem of Human Knowledge: Five Principles

Finally! 🤩 Our position piece: Against the Uncritical Adoption of 'AI' Technologies in Academia:
doi.org/10.5281/zeno...

We unpick the tech industry’s marketing, hype, & harm; and we argue for safeguarding higher education, critical
thinking, expertise, academic freedom, & scientific integrity.
1/n

Congrats Roland :)!

I actually love this!

Exciting to see our protein binder design pipeline BindCraft published in its final form in @Nature ! This has been an amazing collaborative effort with Lennart, Christian, @sokrypton.org, Bruno and many other amazing lab members and collaborators.

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

Thanks! I spoke with Ivan earlier this year about our intersections, I gave a talk on this paper and our pylambdaopt work at the Folding@NYC meeting in Jan. I'm really excited for that as well and generally for your tools. I think they are very powerful and beneficial for the field 💪!

Expanded ensemble was also able to well capture average residue-wise mutation effects, potentially allowing for prediction of beneficial position-wise mutation sites.

We found that, although Flex-ddG was more accurate, this accuracy came from a conservative prediction tendency (predicting most mutations to be neutral). Expanded ensemble however, was better able to predict significantly stabilizing or destabilizing mutations.

Additionally, we re-ran Rosetta-based SSM using the Flex-ddG protocol.

Using Bayesian inference of the convoluted high-throughput FACS data from the publication of these designs [Chevalier, A.; Silva, D. A.; Rocklin, G. J.; Nature 2017, 550, 74–79 doi.org/10.1038/natu... ] we estimated the experimental binding affinities of all site-saturated mutants in the 3 binders.

We used the massive amount of statistics from these calculations to quantify sources of uncertainty. Our uncertainties were distributed bimodally with a mean of ~2 kcal/mol. Sources of larger uncertainties include number of alchemical atoms, charge changes, and slow DOFs.

Massively Parallel Free Energy Calculations for In Silico Affinity Maturation of Designed Miniproteins Computational protein design efforts continue to make remarkable advances, yet the discovery of high-affinity binders typically requires large-scale experimental screening of site-saturated mutant (SS...

Excited to share the final paper from my PhD in @voelzlab.bsky.social , out now in #JCTC @acs.org ! We ran ~43k expanded ensemble free energy calculations on @foldingathome.org to do in silico site saturation mutagenesis on designed hemagglutinin minibinder proteins.

🔗 doi.org/10.1021/acs....