GDTR: Layer-wise Settling Depth Reveals Biological Grammar in Genomic Foundation Models
Cho, Y.; Kang, J.; Park, S.; Kim, S.
Show abstract
Genomic foundation models capture sequence regularities, yet existing interpretability tools rarely ask where in the layer stack a biological grammar becomes stable. We introduce GDTR (Genomic Deep-Thinking Ratio), a training-free residual-stream lens that assigns each nucleotide token a settling depth c(t): the first layer at which its representation stabilises against the post-final-norm reference. On Evo 2 7B, splice donor and acceptor sites settle approximately two layers earlier than intronic contexts, enhancer-like cCREs show a smaller but measurable shift, and a chr22 calibration transfers to held-out chr17. Perturbing canonical splice donors shows that the signal is bidirectional: disrupting the central GT motif deepens settling, whereas shuffling the flanking grammar makes the preserved motif settle earlier. Differential GDTR further reveals consequence-associated peak-disruption depths across ClinVar variants, with synonymous substitutions peaking deepest but with broad class overlap. GDTR therefore provides a layer-wise interpretability axis for genomic foundation models, complementary to existing prediction and variant-scoring tools.
Matching journals
The top 6 journals account for 50% of the predicted probability mass.
Similar papers in this journal
Similar papers in this journal
- Learning a Pairwise Epigenomic and Transcription Factor Binding Association Score Across the Human Genome 93%
- Defining the extent of gene function using ROC curvature 93%
- acmgscaler: An R package and Colab for standardised gene-level variant effect score calibration within the ACMG/AMP framework 93%
"Similar papers" are the closest papers from that journal in the model's embedding space. They show what the match is built on, but the ranking comes mostly from a classifier over the whole training set, not from these examples alone.