Woolfe, A. | bioRxiv Science

bioRxiv2026

Decoding the mystery of ultra-conservation in developmental enhancers: a role for nucleosome positioning, DNA structure and transcription factor binding.

Many human developmental enhancers are characterized by extreme evolutionary constraint in the vertebrate lineage and unique DNA sequence properties, the functional relevance of which is still unknown. Here, we investigate the consequences of their DNA sequence features on three potential aspects important for their function - transcription factor sequence recognition, chromatin accessibility and DNA structure. Using computational predictions in human as well as other vertebrates and invertebrates, we find that conserved non-coding elements (CNEs) are intrinsically nucleosome disfavoring at their core, but favor nucleosome occupancy at their borders driven by distinct nucleotide features conserved over large evolutionary distances. Nevertheless, using genome-wide nucleosome occupancy datasets, we find vertebrate CNEs exhibit higher nucleosome occupancy in comparison to surrounding regions in differentiated cells but a highly accessible conformation in embryonic tissues, suggesting a role for nucleosome positioning in their function. In addition, CNEs are exclusively enriched for homeobox transcription factor motifs, which are found at high density across their sequences. In particular, motifs specifically enriched at the boundary belong to the PBX-HOX, MEIS and POU transcription factor families, known to recognize specific DNA structural features. Consistent with this finding, CNE boundaries are enriched for unusual DNA structural motifs that may constitute a recognition mechanism by transcription factors that bind a narrow minor groove. The finding that extreme nucleotide conservation are likely to be driven by a combination of nucleosome and protein binding constraints provide a potential mechanistic insight into the function of early developmental enhancers.

genomics↗

bioRxiv2025

Paraplume: A fast and accurate paratope prediction method provides insights into repertoire-scale binding dynamics

The specific region of an antibody responsible for binding to an antigen, known as the paratope, is essential for immune recognition. Accurate identification of this small yet critical region can accelerate the development of therapeutic antibodies. Determining paratope locations typically relies on modeling the antibody structure, which is computationally intensive and difficult to scale across large antibody repertoires. We introduce Paraplume, a sequence-based paratope prediction method that leverages embeddings from protein language models (PLMs), without the need for structural input and achieves superior performance across multiple benchmarks compared to current methods. In addition, reweighting PLM embeddings using Paraplume predictions yields more informative sequence representations, improving downstream tasks such as affinity prediction, binder classification, and epitope binning. Applied to large antibody repertoires, Paraplume reveals that antigen-specific somatic hypermutations are associated with larger paratopes, suggesting a potential mechanism for affinity enhancement. Our findings position PLM-based paratope prediction as a powerful, scalable alternative to structure-dependent approaches, opening new avenues for understanding antibody evolution.

bioinformatics↗