Aparajita Karmakar

The paper evaluates feature representations for predicting cancer-associated sORFs in non-coding RNA using traditional machine learning models and finds that k-mer features outperform genomic language model embeddings.

The paper benchmarks siRNA prediction methods, showing that rigorous cross-validation and z-curve representation improve performance and reliability.

This study identifies two potent nanobodies targeting the conserved S2 region of SARS-CoV-2 spike protein that resist viral escape and reduce viral load in a hamster model.