Biology subjects

Didi, K.

Publications and source records attributed to Didi, K..

10 recordsLinked to original sources

bioRxiv2026

Latent generative search unlocks de novo design of untapped biomolecular interactions at scale

De novo protein design has advanced rapidly, yet designing binders to polar, solvent-exposed epitopes and small, flexible ligands remains challenging. Such hydrated surfaces and flexible molecules, including carbohydrates, provide few of the hydrophobic contacts favoured by current methods and have largely resisted de novo binders. To address this challenge, here we introduce latent generative search for binder design, a novel framework that uses reward-guided search at inference time to steer the Proteina-Complexa generative model. The model codesigns sequence and structure - generating them together in a continuous latent space - and thereby removes the inverse-folding step on which current methods rely. In a screen of more than one million designs by multiplexed phage display, latent generative search produced more validated binders than every other method tested, its codesigned sequences surpassing post hoc redesign. It delivered high-affinity binders across therapeutic receptors, a viral attachment protein and intracellular signalling targets. Our approach also accessed previously untapped biology, generating the first de novo proteins that bind a free carbohydrate, including one that discriminates between blood-group antigens - a polar, flexible target class beyond the reach of current design methods.

bioengineering↗

bioRxiv2026

Fast remote nucleotide sequence alignment with Riboseek

Structure prediction has reached RNA, where generating deep alignments is now the bottleneck. We developed Riboseek, a search and alignment tool for RNA and DNA that represents sequences as overlapping di-mers. Riboseek is more sensitive for homology detection than BLASTN and nhmmer, and generates alignments 250- and 376-fold faster than nhmmer and rMSA. With structure-aware realignment, it approaches rMSAs base-pair recovery, while remaining 29-fold faster. We provide a web server and 1.7 million precomputed alignments for putatively structured RNAs. Riboseek is freely available at https://github.com/steineggerlab/riboseek.

bioinformatics↗

bioRxiv2026

AlphaFold Database expands to proteome-scale quaternary structures

Protein function is governed by molecular interactions, yet structural coverage of these interactions remains sparse. The AlphaFold Protein Structure Database (AFDB) transformed access to accurate monomeric protein structures at scale. Here, we expand the AFDB to quaternary structures by predicting 31M candidate homo- and heterodimeric protein complexes, compiled from 4,777 proteomes, including model- and global health organisms. We established confidence criteria via analysis of experimentally determined structures, resulting in 1.81M high-confidence predictions. These models enabled the discovery of emergent structures and topologies not present in monomeric predictions. Additionally, the top 1% of structural clusters accounted for [~]44% of all complexes, and [~]8.3% of clusters were conserved across multiple domains of life, pointing to a substantial fraction of ancient, universally retained assemblies. Structural search highlighted that high-confidence predictions are anchored in experimental multimer space, yet 31.3% extend beyond detectable PDB coverage. These freely accessible proteome-scale predictions facilitate functional and mechanistic hypothesis generation across biology.

synthetic biology↗

bioRxiv2026

Efficient protein structure prediction fromcompact computers to datacenters withOpenFold-TRT

We introduce accelerations for deep learning inference with OpenFold and TensorRT that, combined with MMseqs2-GPU on an x86 system with one NVIDIA RTX PRO 6000 Blackwell Server Edition GPU, reach up to 131x faster inference compared to AlphaFold2. ARM-optimizations enable homology search on power efficient DGX Spark, and allow search beyond available GPU RAM on NVIDIA Grace Hopper Superchip at comparable overall folding speed. These accelerations enable high-throughput protein structure inference at no accuracy cost using familiar tools.

bioinformatics↗

bioRxiv2026

FLIP2: Expanding Protein Fitness Landscape Benchmarks for Real-World Machine Learning Applications

Machine learning methods that predict protein fitness from sequence remain sensitive to changes in data distributions, limiting generalization across common conditions encountered in protein engineering. Practically, protein engineers are thus left wondering about the effective utility of ML tools. The FLIP benchmark established protocols for testing generalization under some domain shifts, but it was limited to measurements of thermostability, binding, and viral capsid viability. We introduce FLIP2, a protein fitness benchmark spanning seven new datasets, including enzymes, protein-protein interactions, and light-sensitive proteins, as well as splits that measure generalization relevant to real-world protein engineering campaigns. Evaluating a suite of benchmark models across these datasets and splits reveals that simpler models often matched or outperformed fine-tuned protein language models on FLIP2, challenging the utility of existing transfer learning techniques. Provenance for all datasets has been recorded and we redistribute all data CC-BY 4.0 to facilitate continued progress.

bioengineering↗

bioRxiv2025

De novo Design of All-atom Biomolecular Interactions with RFdiffusion3

Deep learning has accelerated protein design, but most existing methods are restricted to generating protein backbone coordinates and often neglect interactions with other biomolecules. We present RFdiffusion3 (RFD3), a diffusion model that generates protein structures in the context of ligands, nucleic acids and other non-protein constellations of atoms. Because all polymer atoms are modeled explicitly, conditioning the model on complex sets of atom-level constraints for enzyme design and other challenges is both simpler and more effective than previous approaches. RFD3 achieves improved performance compared to prior approaches on a range of in silico benchmarks with one tenth the computational cost. Finally, we demonstrate the broad applicability of RFD3 by designing and experimentally characterizing DNA binding proteins and cysteine hydrolases. The ability to rapidly generate protein structures guided by complex sets of atom-level constraints in the context of arbitrary non-protein atoms should further expand the range of functions attainable through protein design.

biochemistry↗

bioRxiv2025

Accelerating Biomolecular Modeling with AtomWorks andRF3

Deep learning methods trained on protein structure databases have revolutionized biomolecular structure prediction, but developing and training new models remains a considerable challenge. To facilitate the development of new models, we present AtomWorks: a broadly applicable data framework for developing state-of-the-art biomolecular foundation models spanning diverse tasks, including structure prediction, generative protein design, and fixed backbone sequence design. We use AtomWorks to train RosettaFold-3 (RF3), a structure prediction network capable of predicting arbitrary biomolecular complexes with an improved treatment of chirality that narrows the performance gap between closed-source AlphaFold3 (AF3) and existing open-source implementations. We expect that AtomWorks will accelerate the next generation of open-source biomolecular machine learning models and that RF3 will be broadly useful as a structure prediction tool. To this end, we release the AtomWorks framework (https://github.com/RosettaCommons/atomworks), together with curated training data, code and model weights for RF3 (https://github.com/RosettaCommons/modelforge) under a permissive BSD license.

biochemistry↗

bioRxiv2024

GPU-accelerated homology search with MMseqs2

Rapidly growing protein databases demand faster sensitive sequence similarity detection. We present GPU-accelerated search utilizing intra-query parallelization delivering 6x faster single-protein searches compared to state-of-the-art CPU methods on 2x64 cores--speeds previously requiring large protein batches. It is most cost effective, including in large-batches at 0.45x MMseqs2-CPU speed (8 GPUs delivering 2.4x). It accelerates ColabFold structure prediction 31.8x compared to AlphaFold2 and Foldseek search 4-27x. MMseqs2-GPU is open-source at mmseqs.com.

bioinformatics↗

bioRxiv2023

BioModelsML: Building a FAIR and reproducible collection of machine learning models in life sciences and medicine for easy reuse

Machine learning (ML) models are widely used in life sciences and medicine; however, they are scattered across various platforms and there are several challenges that hinder their accessibility, reproducibility and reuse. In this manuscript, we present the formalisation and pilot implementation of community protocol to enable FAIReR (Findable, Accessible, Interoperable, Reusable, and Reproducible) sharing of ML models. The protocol consists of eight steps, including sharing model training code, dataset information, reproduced figures, model evaluation metrics, trained models, Dockerfiles, model metadata, and FAIR dissemination. Applying these measures we aim to build and share a comprehensive public collection of FAIR ML models in the BioModels repository through incentivized community curation. In a pilot implementation, we curated diverse ML models to demonstrate the feasibility of our approach and we discussed the current challenges. Building a FAIReR collection of ML models will directly enhance the reproducibility and reusability of ML models, minimising the effort needed to reimplement models, maximising the impact on the application and significantly accelerating the advancement in the field of life science and medicine.

bioinformatics↗

bioRxiv2023

AbNatiV: VQ-VAE-based assessment of antibody and nanobody nativeness for engineering, selection, and computational design

Monoclonal antibodies have emerged as key therapeutics, and nanobodies are rapidly gaining momentum following the approval of the first nanobody drug in 2019. Nonetheless, the development of these biologics as therapeutics remains a challenge. Despite the availability of established in vitro directed evolution technologies that are relatively fast and cheap to deploy, the gold standard for generating therapeutic antibodies remains discovery from animal immunization or patients. Immune-system derived antibodies tend to have favourable properties in vivo, including long half-life, low reactivity with self-antigens, and low toxicity. Here, we present AbNatiV, a deep-learning tool for assessing the nativeness of antibodies and nanobodies, i.e., their likelihood of belonging to the distribution of immune-system derived human antibodies or camelid nanobodies. AbNatiV is a multi-purpose tool that accurately predicts the nativeness of Fv sequences from any source, including synthetic libraries and computational design. It provides an interpretable score that predicts the likelihood of immunogenicity, and a residue-level profile that can guide the engineering of antibodies and nanobodies indistinguishable from immune-system-derived ones. We further introduce an automated humanisation pipeline, which we applied to two nanobodies. Wet-lab experiments show that AbNatiV-humanized nanobodies retain binding and stability at par or better than their wild type, unlike nanobodies humanised relying on conventional structural and residue-frequency analysis. We make AbNatiV available as downloadable software and as a webserver.

bioengineering↗