proomt

Search

Search posts, papers, and topics

All posts

Microsoft ResearchFelix Pultar, John Gardner, Guoqing Liu, Marwin Segler5 min readintermediate

Improving synthesis prediction of small molecules at scale with RetroChimera

Read on microsoft.com

Summary

RetroChimera is an open‑source retrosynthesis system that ensembles a Transformer‑based de‑novo model and a GNN template selector with a learned ranking layer. In blind expert evaluations it achieved a 90 % route acceptance rate, far above prior models.

  • RetroChimera combines R‑SMILES 2 (Transformer) and NeuralLoc (GNN) using a learned voting scheme to rank predictions.
  • The ensemble matches the stronger sub‑model per reaction class, improving recall of rare reaction types.
  • Expert chemists accepted 90 % of full routes from RetroChimera versus 20‑50 % for baseline models.
  • The code, pretrained weights, and Azure Foundry access are released under an MIT license.

Chemists and ML engineers building synthesis planning pipelines should care because RetroChimera offers higher-quality routes and an open‑source, extensible framework for scaling retrosynthesis.

7/10
machine learningtransformersgnnchemistryretrosynthesis

Related reading

  1. Advancing AI for biology: Teaching models to design and characterize antibodies

    Amazon Bio Discovery developed three AI models: MochiBind for fast, sequence-based antibody binding ranking, CA-MAP for context-aware developability prediction robust to batch effects, and an agent-guided system for de novo antibody design. These advancements aim to accelerate and improve the accuracy of antibody drug discovery, with experimental validation for a novel cancer target.

    Amazon Scienceamazon.science10 min

  2. HypoEvolve: Genetic Algorithms Enable Multi-Agent LLMs to Discover Scientific Hypotheses

    The paper presents HypoEvolve, a generational genetic algorithm that coordinates specialized LLM agents to iteratively propose, critique, and refine scientific hypotheses. On a drug‑repurposing benchmark across 34 cancer types, it outperforms six baselines, achieving a DepMap selectivity of 0.171 versus 0.115.

    Hugging Face Daily Papersarxiv.org1 minpaper

  3. JEPA-Anything: Learning Predictive Models across Different Worlds

    JEPA-Anything extends joint‑embedding predictive architectures with orthogonal predictive factorization, letting a single model learn complementary latent factors that can be recombined for prediction across disparate domains. The paper shows consistent performance gains on ten dynamics tasks, molecular simulations, and clinical event forecasting, plus experimental validation of a biologically‑de…

    Hugging Face Daily Papersarxiv.org1 minpaper

  4. Grounded Skill Synthesis from Code at Scale for Agentic Intelligence

    Code2Skill is an automated pipeline that mines popular GitHub repositories to extract verifiable, implementation‑anchored procedural “skills”. It builds a bank of ~1 M skill records (atomic ops, workflows, patterns) with provenance metadata, verifies each via blind reconstruction, and shows that augmenting LLM‑based agents with these skills yields an average 11.7% performance lift across 72 proto…

    Hugging Face Daily Papersarxiv.org1 minpaper

  6. Reverse-engineered Jev-like model

    Jevlike is an open‑source starter model that scores a list of text options in a single forward pass. It provides a minimal architecture (option queries, shared dot‑product scorer), synthetic data generation, training/evaluation CLI, and examples on Doom and chess. The repo supports a byte‑level encoder or a frozen Hugging‑Face encoder (e.g., Qwen2.5‑0.5B), runs on CPU/MPS/CUDA, and reports benchm…

    Hacker News front pagegithub.com4 minreleaseHN16224