Biology subjects

de Polavieja, G.

Publications and source records attributed to de Polavieja, G..

6 recordsLinked to original sources

bioRxiv2025

C.La.P.: Enhancing transformer-based genomic signal modeling by integrating DNA sequences and chromatin accessibility data

Transformers have shown promise in chromatin modeling but have primarily relied solely on reference DNA sequences, limiting their utility across multiple biological contexts. In this work we enhance transformer models by integrating reference sequences with ATAC-seq, a chromatin accessibility assay. Our Chromatin LAnguage Processing (CLaP) model combines a convolutional tokenizer, a transformer encoder, and task-specific components to predict multiple genomic signals from a single input. After pre-training with masked nucleotide prediction, CLaP achieved per-token F1 scores exceeding 0.8 for three target ChIP-seq assays in fine-tuning. Attention mechanism analysis revealed that CLaP detects CTCF binding sites with nucleotide-level precision by learning the sequence preference of the CTCF factor and the characteristic ATAC-seq patterns that are caused by protein binding events. Additionally, CLaP predicts protein-DNA binding events not captured by the ChIP-seq ground truth. These findings highlight CLaPs potential to expand chromatin modeling by incorporating molecular assay data alongside sequence information.

bioinformatics↗

bioRxiv2024

Phenotypic impact of individual conserved neuronal microexons and their master regulators in zebrafish

Microexons exhibit striking evolutionary conservation and are subject to precise, switch-like regulation in neurons, orchestrated by the splicing factors Srrm3 and Srrm4. Disruption of these regulators in mice leads to severe neurological phenotypes, and their misregulation is linked to human disease. However, the specific microexons involved in these phenotypes and the effects of individual microexon deletions on neurodevelopment, physiology, and behavior remain poorly understood. To explore this, we generated zebrafish lines with deletions of 18 individual microexons, alongside srrm3 and srrm4 mutant lines, and conducted comprehensive phenotypic analyses. We discovered that while loss of srrm3, alone or together with srrm4, resulted in significant alterations in neuritogenesis, locomotion, and social behavior, individual microexon deletions typically produced mild or no noticeable effects. Nonetheless, we identified specific microexons associated with defects in neuritogenesis (evi5b, vav2, itsn1, src) and social behavior (vti1a, kif1b). Additionally, most microexon deletions triggered coordinated transcriptomic changes in neural pathways, suggesting the presence of molecular compensatory mechanisms. Our findings suggest that the severe phenotypes caused by Srrm3/4 depletion arise from the combined effects of multiple subtle disruptions across various cellular pathways, which are individually well-tolerated.

molecular biology↗

bioRxiv2018

Sensory cheating: adversarial body patterns can fool a convolutional visual system during signaling

Animals often assess each other by paying special attention to signals, which help to communicate the quality of each individual. When there is a conflict of interest between the signaler and the receiver, then the signaler has an incentive to cheat by producing signals which exaggerate its apparent quality. One opportunity for cheating might be to rely on sensory illusions, but it has been difficult to study sensory cheating because we have lacked quantitative models of complex visual perception. Here we address this problem by taking advantage of recent advances in modeling visual brain areas as convolutional neural networks. Given these models, we use the technique of adversarial perturbations to show how sensory cheating can shape animal appearance while nevertheless resulting in an evolutionarily stable signaling system. In our simulations, animals typically evolve exaggerated color patterns which might be analogous to the evolution of colorful body patterns in guppies.

neuroscience↗

bioRxiv2018

idtracker.ai: Tracking all individuals in large collectives of unmarked animals

Our understanding of collective animal behavior is limited by our ability to track each of the individuals. We describe an algorithm and software, idtracker.ai, that extracts from video all trajectories with correct identities at a high accuracy for collectives of up to 100 individuals. It uses two deep networks, one detecting when animals touch or cross and an-other one for animal identification, trained adaptively to conditions and difficulty of the video.

animal behavior and cognition↗

bioRxiv2018

Towards a unification of niche and neutral models of community ecology

Ecological models of community dynamics fall into two main categories. The neutral theory of biodiversity correctly predicts various large-scale ecosystem characteristics such as the species abundance distributions. On a smaller scale, the niche theory of species competition explains population dynamics and interactions between two to a dozen species. Despite the successes of the two theories, they rely on two contradictory assumptions. In the neutral theory each species is competitively equivalent while in the niche theory every species is specialized to exploit a specific part of its environment. Here we propose a resolution to this contradiction using a game theory model of competition with an attractor hyperplane as its equilibrium solution. When the population dynamics shifts within the hyperplane, it is selectively neutral. However, any movement perpendicular to the hyperplane is subject to restoring forces similar to what is predicted by the niche theory. We show that this model correctly reproduces empirical species abundance distributions and is also compatible with species removal experiments.

ecology↗

bioRxiv2017

Opponent assessment and conflict resolution through mutual motor coordination

Most animals fight by repeating complex stereotypic behaviors, yet the internal structure of these behaviors has rarely been dissected in detail. We characterized the internal structure of fighting behaviors by developing a machine learning pipeline that measures and classifies the behavior of individual unmarked animals on a sub-second timescale. This allowed us to quantify several previously hidden features of zebrafish fighting strategies. We found strong correlations between the velocity of the attacker and the defender indicating a dynamic matching of approach and avoidance efforts consistent with the cumulative assessment model of aggression. While velocity matching was ubiquitous, the spatial dynamics of attacks showed phase-specific differences. Contest phase attacks were characterized by a paradoxical side-ways attraction of the retreating animal towards the attacker suggesting that the defender combines avoidance maneuvers with display maneuvers. Post-resolution attacks lacked display-like features and the the defender was avoidance-focused. From the perspective of the winner, game theory modeling further suggested that highly energetically costly post resolution attacks occurred because the winner was trying to increase its relative dominance over the loser. Overall, the rich structure of zebrafish motor coordination during fighting indicates a greater complexity and layering of strategies than has previously been recognized.

animal behavior and cognition↗