Ogün Adebali

Evo2 needs A100/H100 GPUs. GPN-MSA: 86M parameters. PHACTn: standard CPUs, no training, no GPU. ~27 CPU-hours for 100 × 10kb windows at 50 parallel jobs. PHACTn provides phylogeny-aware variant scoring for the entire genome.

We built a "hard-case" dataset: variants where existing tools disagree. The toughest 9% of the benchmark. PHACTn AUROC: 0.921 Evo2-7B: 0.884 GPN-MSA: 0.866 When tools fight, PHACTn often gets it right.

Non-coding is hardest. PHACTn AUROC: 0.805 🥇 GPN-MSA: 0.787 Evo2-7B: 0.717 And on the TraitGym Mendelian benchmark (113 diseases from OMIM): PHACTn AUROC: 0.914, AUPR: 0.736 — top of the leaderboard.

On ~294k variants (ClinVar pathogenic + gnomAD common): PHACTn AUROC: 0.988

PHACTn: - Reconstructs ancestral probabilities at every tree node - Corrects for gaps with binary ASR - Traverses the tree from query to outward - Weights substitution events inversely by phylogenetic distance - Applies a sigmoid gap-frequency correction With only 4 parameters!

Standard ancestral reconstruction ignores gaps and treats them as missing data, forces probabilities to sum to 1 across A/C/T/G. This creates "ghost" signal at positions dominated by deletions. PHACTn uses a dual binary+nucleotide ASR approach to correct for this explicitly.

Two alignments might look identical by nucleotide frequency, yet tell different stories. One position may have switched three times independently: strong tolerance indicator. Another may have done it once: weaker tolerance. PHACTn traverses the tree to distinguish these cases.

We're excited to introduce the GPCR Evolution Database - an open-access resource for evolutionary analysis of GPCRs 🎉 👉 gpcrevolution.org 📄 doi.org/10.64898/2026.09.08.… 🧵

Evolution has already explored an enormous sequence space. By combining conservation within orthologs and divergence across paralogs, we can identify where proteins preserve function—and where evolution creates innovation. Paper: doi.org/10.1111/bph.70538

The framework also uncovered family-specific evolutionary signatures: • unique extracellular organization in olfactory receptors • ICL2 specialization in bitter taste receptors • conserved cholesterol-contact positions • cross-family structural motifs