Cross-Species Antimicrobial Resistance Prediction from Genomic Foundation Models

TL;DR

This paper addresses the challenge of predicting antimicrobial resistance across different bacterial species by developing representation strategies that improve cross-species generalization of genomic models.

Contribution

It introduces diagnostics for selecting stable model layers and a local activation pattern aggregation method that enhances cross-species AMR prediction.

Findings

Layer stability diagnostics identify optimal embedding layers.

Local activation pattern aggregation improves cross-species generalization.

Preserving localized signals enhances resistance prediction accuracy.

Abstract

Cross-species antimicrobial resistance (AMR) prediction is fundamentally an out-of-distribution (OOD) generalization problem: models trained on one set of bacterial taxa must transfer to phylogenetically distinct genomes that may rely on different resistance mechanisms. Across species, resistance arises from a heterogeneous mixture of localized, horizontally transferred gene cassettes and diffuse species-specific genomic backgrounds, making successful transfer inherently mechanism-dependent. Using a strict species holdout protocol, we first establish an interpretable k-mer baseline with Kover and show that strong within-species performance collapses under true cross-species evaluation. This motivates representation-level approaches that preserve transferable biological signals rather than amplify phylogenetic shortcuts. We investigate genomic foundation model embeddings derived from Evo-1-8k-base and introduce diagnostics for layer selection based on activation scale, isotropy, effective rank, and cross-seed stability under native bfloat16 inference. These analyses identify a stability boundary in deeper layers and reveal that embeddings extracted near this boundary provide more robust representations for downstream prediction. To preserve localized resistance signals, we treat per-window embeddings as an ordered multivariate signal and apply MiniRocket to summarize multi-scale local activation patterns instead of relying on global pooling. Our results show that aggregation strategy plays a central role in cross-species AMR prediction and that preserving local activation patterns substantially improves generalization when resistance mechanisms are localized.