Alexander H. Schug

This paper introduces RNA-MDLM, a model for generating and inpainting non-coding RNA sequences, showing it produces sequences with natural composition and folding statistics.

The paper uses high-resolution vascular data from mouse brains to simulate tumor growth, finding vessel density strongly influences growth rate and enabling brain-wide tumor predictions.

The paper evaluates how contact map topology affects RNA structure prediction and introduces a Gaussian score to assess contact distribution quality.

NucleicBERT is a self-supervised masked-language model for RNA that learns contextual representations without alignments or evolutionary data, outperforming existing models in structural and functional tasks.

NucleoSeeker curates high-quality RNA structure datasets by filtering the PDB database, improving dataset quality for training and evaluating RNA structure prediction methods.

CoCoNet improves RNA contact map prediction using a coevolutionary model and convolutional neural network, achieving 70% higher accuracy than DCA on 60 RNA structures.

The paper evaluates DCA-based methods for RNA contact prediction using a curated dataset of RNA structures and finds minor performance differences between methods.

pydca is a Python software for Direct Coupling Analysis of RNA and protein sequences, offering tools for sequence alignment, contact map visualization, and coevolutionary coupling inference.