Altuna Akalin

This study shows that oncogene inactivation-induced senescence both limits tumor growth and promotes relapse by creating a pro-recurrence microenvironment.

The paper trains a supervised VAE on bulk RNA-seq data to learn a unified tissue-aware representation, achieving high tissue classification accuracy and demonstrating the impact of dataset scale.

This study uses large language models to extract genomic biomarkers from oncology trials, showing open-source models outperform closed-source ones and benefit from fine-tuning.

This paper introduces mergen, an R package using LLMs to automate data analysis by generating and executing code based on user descriptions, showing promise but highlighting challenges in complex tasks.

scMaui is a deep learning framework that integrates single-cell multiomics data, handles batch effects, and addresses missing data, showing improved performance in downstream analyses.

Flexynesis provides a versatile deep learning framework for integrating bulk multi-omics data, offering usability, adaptability, and benchmarking across diverse datasets.

This study shows that a single cocaine exposure causes long-lasting 3D genome structure changes in midbrain dopamine neurons, affecting gene regulation and synaptic plasticity.

This study shows DNA virus infections activate transposable elements (TEs) through KAP1/TRIM28 inhibition and DUX4 expression, with in vivo correlations in cancer patients.

The paper investigates how pathogenic mutations at phosphorylation sites affect protein-protein interactions, showing that such mutations disrupt phosphorylation-dependent motifs and alter cellular functions.

VoltRon is an R package for spatial omics analysis that integrates multi-resolution and multi-omics data through image registration and spatial alignment.

This study shows that combining transcriptome data with gene panel features improves drug response prediction in cancer.

This paper introduces a variational auto-encoder for single-cell ATAC-seq data that performs dimensionality reduction and batch correction simultaneously.

This study uses deep learning to analyze multi-omics data and identify cancer-related biomarkers that can classify cancer types, predict survival, and guide clinical decisions.

The paper functionally analyzes regulatory sequences in C. elegans by introducing mutations and measuring their effects on gene expression and fitness.

The paper uses bulk and single-cell transcriptomics to analyze SARS-CoV-2 infection in human cell lines, identifying HSP90 as a potential therapeutic target.

The paper evaluates epigenetic methods using reference materials, finding high concordance between assays and differences in performance across platforms.

The paper uses snRNAseq to identify nuclear subtypes in skeletal muscle cells, revealing new compartments and functional signatures in dystrophic and regenerating muscle.

Janggu is a Python library that simplifies deep learning in genomics by providing reusable data tools and model evaluation features, improving performance through high-order sequence features.

The paper functionally annotates human long non-coding RNAs by systematically knocking down 285 lncRNAs and analyzing molecular phenotypes using CAGE sequencing.

The paper uses sci-ATAC-seq and a new tool, ScregSeg, to analyze chromatin dynamics in zebrafish embryos, revealing roles of npas4l/cloche in cell and regulatory states.

The paper investigates cell-type specific 3D chromatin structures in the brain using immunoGAM, finding TAD melting, neuron-specific contacts, and compartmentalization patterns related to gene expression and neurodegenerative diseases.

This paper develops a deep learning method to classify colorectal cancer subtypes and match them to cell lines using multi-omics data.

The paper uses single-cell RNA-sequencing to study HSV-1 infection in human fibroblasts, identifying NRF2 activation as an antiviral response that reduces virus production.

The paper investigates whether HOT regions are genuine transcription factor binding sites or ChIP-seq artifacts, finding they likely result from technical biases and specific sequence features.

The paper identifies functional miRNA::mRNA interactions in Drosophila using AGO CLIP experiments and quantifies their impact on gene expression and translation.

The paper analyzes RBP-RNA interactions across 64 RBPs to identify regulatory modules and their functional roles in RNA metabolism and localization.

The paper presents reproducible genomics analysis pipelines using GNU Guix for RNA-seq, ChIP-seq, Bisulfite-seq, and single-cell RNA-seq, ensuring consistent results across different environments.

The paper identifies FACT as a conserved reprogramming barrier in C. elegans and humans that maintains cell identity by modulating chromatin accessibility.

netSmooth improves scRNA-seq clustering by smoothing expression values using network diffusion and gene expression priors.

This paper reviews techniques and software for analyzing bisulfite sequencing data, focusing on quality checks, differential methylation, segmentation, and annotation methods.