Research

Model Performance Series

June 2026

Astra AI on Ion Channels

Ion channels are among the most consequential — and most treacherous — protein classes in drug discovery. They are validated targets across pain, epilepsy, and arrhythmia; and one of them, hERG, is the most prominent anti-target in small-molecule development.

Çağlar Bozkurt, Aniruddh Goteti · Orbion GmbH · Benchmarked on 2,700 Ion Channels

0.97
AUROC · Residue-Level Topology
95.6%
GO Top-5 Function Recall
56%
Ligand-Identity Recall

What We Found

Three Results That Matter

Each section of the whitepaper reports the headline performance for one prediction area, its known weaknesses, and where we recommend it for production use.

01

0.97 AUROC on Residue-Level Topology

Per-residue predictions agree with UniProt transmembrane annotations (AUPRC 0.91, F1 0.87, n = 307). The disorder model reaches AUROC 0.89.

02

F1 Up to 0.90 on PTM Sites

Strongest on disulfide bonds (0.90) and N-linked glycosylation (0.88). All 39 modification classes covered, with two operating points reported.

03

56% Ligand Recall — Pocket-Level Triage

The aggregate masks a clean split. On ligand-gated channels the model localizes the pocket well. On voltage-gated channels — including the hERG safety case — it recovers the ligand identity, but the pore-block site is structurally diffuse. Read it as a ligand hypothesis set, not a residue-level contact map.

Every number above is reproduced in the whitepaper against public references — Swiss-Prot annotation, PDB co-crystal contacts, and curated experimental thermal-shift data. We report where the models are weak as plainly as where they are strong.

Read the Whitepaper