Follow research directions that could alter real discovery decisions—from cryptic pockets and induced-fit co-folding to virtual cells and genome-scale design. BioAtlas keeps the opportunity and the uncertainty together.
Can the community reproduce and extend proprietary all-atom drug-design engines with open training code, checkpoints and benchmarks?
Can models forecast how cell populations respond to unseen drugs, gene edits, cytokines and environmental changes across biological contexts?
Can one model predict both the bound geometry and the energetic ranking needed to prioritize compounds?
Can a foundation model read, predict and design biological sequence continuously from single nucleotides to megabase-scale genomes?
Can a model reveal ligandable pockets that are hidden in the unbound protein and only open after a ligand or allosteric change?
Can structure models represent large ligand-driven protein rearrangements when the target, pocket or conformational transition is far from training examples?
Can general co-folding models accurately resolve antibody–antigen and other biologic interfaces with low sequence homology?
Can virtual screening retain useful structural and affinity accuracy without expensive all-atom diffusion for every compound?
Can a single model predict how coding and non-coding variants alter expression, splicing, chromatin and regulatory binding over long genomic context?
Can one generative model reason jointly over protein sequence, structure and function and create functional proteins from mixed prompts?
Can AI generate high-affinity protein binders for diverse targets with fewer rounds of experimental optimization?
Can RNA programmably specify both target and donor DNA to insert, excise or invert large sequences without relying on conventional CRISPR cutting and repair?
Can the community reproduce and extend proprietary all-atom drug-design engines with open training code, checkpoints and benchmarks?
OpenDDE proposes a shared all-atom co-folding foundation with released training and inference pipelines, checkpoints, data processing and scaling analyses.
Open implementations make replication, ablation, safety review, fine-tuning and domain-specific extension possible instead of limiting progress to hosted inference.
OpenDDE is a very recent July 2026 preprint. Its claimed parity has not yet received broad independent evaluation.