Vasileios Konstantakos

DeepSCENIC uses transfer learning from sequence-to-function models to infer causal gene regulatory networks from single-cell multiome data.

The paper introduces CPP2Vec, a Word2Vec-based method for predicting cell-penetrating peptides, and compares it with pre-trained LLM-based models, showing strong predictive performance.

This study identifies a lipid-mediated signaling pathway driven by microglia that promotes glioblastoma invasion and shows that targeting this pathway can suppress tumor invasion.

The paper identifies cis-acting genetic variants affecting enhancer function in Parkinson's Disease using multi-omics data and develops models to predict their regulatory impact.

CREsted is a software package for modeling and designing cell type-specific enhancers using deep learning, validated across tissues and species.

The paper evaluates computational methods for predicting cell type-specific enhancers in the mouse cortex using in vivo validation data and finds that open chromatin and sequence models improve prediction accuracy.

CRISPRedict provides accurate and interpretable predictions for CRISPR-Cas9 guide RNA efficiency using a simple linear model, outperforming some existing tools and showing robust performance across different datasets.