Computational approaches using machine learning, deep learning, and foundation models to accelerate drug discovery — from target identification and molecular design to ADMET prediction and clinical trial optimization.
MIT's open-weight model matches AlphaFold3 on protein-ligand complexes while running on a single GPU. For computational chemistry teams, the implications for hit-to-lead timelines are significant — and immediate.
Apr 15, 2026 · 12 min read
The AI-designed PI3Kα inhibitor shows durable responses in HR+/HER2- breast cancer with mutant-selective specificity. A landmark for computationally designed oncology drugs.
Apr 14, 2026 · 6 min read
The largest Series B in AI pharma history signals growing conviction that foundation models can compress timelines from target to IND.
Apr 13, 2026 · 4 min read
In a landmark deal structure for AI drug discovery, Lilly commits a mid-eight-figure annual license fee to Chai Discovery for access to its biologics design foundation model. Separately, GSK pays $50M upfront to NOETIK for its oncology model. The era of AI-as-infrastructure-for-pharma is here.
Apr 12, 2026 · 7 min read
Targeted protein degradation meets deep learning. Where the biology is ready, where the data is not, and what it means for drug programs targeting tau and alpha-synuclein.
Apr 12, 2026 · 9 min read
The updated diffusion model achieves 58% top-1 success rate on PoseBusters, narrowing the gap with physics-based methods while maintaining speed advantages.
Apr 11, 2026 · 7 min read
An industry-wide analysis finds AI-first drug candidates are achieving Phase I success rates roughly double the historical 50% average. With fewer than 15 AI-first candidates completing Phase II and none yet FDA-approved, 2026 becomes the year the hype meets definitive clinical validation.
Apr 11, 2026 · 6 min read
The collaboration aims to run millisecond-scale MD simulations for thousands of compounds simultaneously, a potential inflection point for free energy perturbation workflows.
Apr 10, 2026 · 5 min read
A new training framework for multimodal AI shows that domain-specific foundation models outperform general-purpose models by an order of magnitude on key drug discovery tasks. General LLMs fell short on approximately half the tested benchmarks, reinforcing that biology needs purpose-built models.
Apr 10, 2026 · 5 min read
The GENESIS-IPF trial delivers the first clinical proof-of-concept for a fully AI-discovered drug: rentosertib, a TNIK inhibitor designed by Insilico Medicine's Chemistry42 platform, demonstrates dose-dependent FVC improvement in idiopathic pulmonary fibrosis patients. Phase IIb is underway.
Apr 10, 2026 · 8 min read
A mid-stage biotech credits the open-source ADMET prediction tool with flagging a metabolic liability that traditional assays missed until later stages.
Apr 9, 2026 · 5 min read
Protein folding benchmarks shift again, two more AI-native biotechs enter the clinic, and why retrosynthesis models still struggle with stereochemistry.
Apr 7, 2026 · 8 min read
Predicts pharmacokinetics, drug-likeness, and medicinal chemistry friendliness. Known for the BOILED-Egg visualization and Bioavailability Radar.
Open-source D-MPNN library for molecular property prediction. The architecture underlying ADMET-AI and ADMETlab 3.0. Used in Halicin antibiotic discovery.
Commercial platform predicting 175+ ADMET properties with integrated PBPK (GastroPlus) and generative drug design (AIDD module).
Predicts PK and physicochemical properties based on full 3D molecular structure. Part of the Schrödinger Drug Discovery Platform.
Equivariant ML force field for organic molecules covering H, C, N, O, F, P, S, Cl, Br, I (~90% of drug-like space). Near-DFT accuracy for torsion profiles.
Gold standard drug knowledge resource. 4,563 FDA-approved drugs, 6,231 investigational drugs, 1.4M drug-drug interactions, comprehensive target annotations.