Search

Drug discovery with NVIDIA and cutting-edge AI: creating pharmaceutical synthesis enzymes from scratch

Translated title. Original: エヌビディアと最先端AIによる創薬：ゼロから医薬品合成酵素を生み出す

A method named Proteina-Complexa performs joint sequence-structure design in a continuous latent space with inference-time search for de novo protein binders. In a large benchmark with 1M proteins across 127 targets via multiplexed phage display, it surpasses open methods, delivering picomolar binders and >50% hit rates.

3D modeling and AI redefine drug discovery

Translated title. Original: Modelado 3D e IA redefinen el descubrimiento de fármacos

AI is reshaping drug discovery by enabling large-scale protein-structure prediction and virtual screening, speeding candidate generation and reducing lab work; however, real-world failures still attribute fault to algorithms, while startups push simulation-driven pipelines.

Google launches Cheminformatics Skills for Antigravity: Agent-based new drug design with the Crizotinib case (feat. RDKit, ChEMBL) - Reading & Information Sharing - PyTorchKR

Translated title. Original: Google, Antigravity용 Cheminformatics Skills 출시: 크리조티닙 사례로 본 에이전트 신약 설계 (feat. RDKit, ChEMBL) - 읽을거리&정보공유 - PyTorchKR

The document proposes speeding up agent-based scientific workflows by improving grounding and token efficiency, and by integrating insights from multiple databases and tools such as AlphaGenome, AFDB, and UniProt.

Antigravity: chemoinformatics by agents | NeoTeo

Translated title. Original: Antigravity : la chimioinformatique par agents | NeoTeo

Google DeepMind presents Antigravity's chemistry workflow, where a language model guides research and determinist tools perform molecular calculations, using RDKit with ChEMBL and PubChem; demonstrations show curated compounds, synthesis routes, and evaluation considerations, with restricted enterprise access and six-month free trial for eligible scientists.

Antigravity: this is how Google's chemo-informatics works | NeoTeo

Translated title. Original: Antigravity: así funciona la quimioinformática de Google | NeoTeo

Google DeepMind introduces Cheminformatics Skills for Antigravity, where a language model coordinates workflow and deterministic chemistry tools perform molecular calculations; proposed molecules require synthesis and experimental validation, with 568 compounds cured demonstrated, and methods to assess generalization across chemical families. Access is offered via the Antigravity app and Science add-on, with tiered eligibility for researchers and enterprise accounts.

[2610.39714] MoLRSI: A Recursive Self-Improving AI agent for Molecular Discovery

MoLRSI is an AI agent that iteratively improves molecular discovery workflows and models using cross-task memory, achieving higher hit rates and validated binders in antibiotic and peptide–MHC discovery compared with prior methods.

Job Application for Research Leader / Senior Research Investigator (Computational Drug Design), London at Isomorphic Labs

Isomorphic Labs is hiring senior computational chemistry researchers to lead early-stage hit-finding or drug-design work across active programs. The roles combine AI-based drug discovery, cross-team project leadership, and, for some hires, management of a small team.

Where Human Insight Meets Artificial Intelligence - Evotec

Evotec combines AI with biology and experimental design to accelerate end-to-end drug discovery and development, leveraging high-throughput screening, multiomics, and in silico design. The company emphasizes human expertise as central to integrating AI into its workflows.

AI/ML Now: Rethinking Retention Time Prediction using Machine Learning | LCGC International

Sebastien Böcker discusses a machine learning method that predicts small-molecule retention times across laboratory systems. It predicts retention order, then uses a small set of anchor compounds to convert that order into times without retraining.

Researchers Map How Dipoles Tune Polaron Binding Energy

Researchers apply a functional determinant approach to compute exact impurity spectra in dipolar gas systems, revealing ground states with both even- and odd-wave bound states whose occupancy is tunable by dipole length and which weaken Anderson’s orthogonality catastrophe; the work, conducted with ETH Zürich and Universität Heidelberg, advances understanding of polaron formation in ultracold dipolar molecules beyond single-excitation models.