Pan-microalgal dark proteome mapping via interpretable deep learning and synthetic chimeras.

Nelson, David R; Jaiswal, Ashish Kumar; Samir Ismail, Noha; Mystikou, Alexandra; Salehi-Ashtiani, Kourosh · Patterns (N Y) · 2025

basic_science · Level V

Where this comes from

Abstract

Microalgal genomes contain a vast "dark proteome"-sequences lacking detectable homology that evade conventional classification tools. We developed LA<sup>4</sup>SR (language modeling with AI for algal amino acid sequence representation), a framework using transformer- and state-space models to classify translated ORFeomes across ten algal phyla. Training on ∼77 million sequences, LA<sup>4</sup>SR achieves near-complete recall, accelerates classification by ∼10,701× relative to BLASTP<sup>+</sup>, and generalizes robustly to unseen sequences using less than 2% of available data. Models trained on synthetic, chimeric (terminal information [TI]-free) sequences maintained high accuracy, demonstrating that internal sequence features alone can drive robust classification. Inference speed and scalability were further enhanced under TI-free settings, supporting rapid annotation of large proteomic datasets. Custom explainability tools revealed interpretable amino acid patterns linked to evolutionary and biophysical features. Designed for accessibility across disciplines, LA<sup>4</sup>SR integrates biological context and computational innovation in parallel, enabling both biologists and data scientists to interrogate the microbial dark proteome.