Melcy Philip

The paper develops a targeted reference database, AQUAeD-DB, to improve taxonomic accuracy of 16S rRNA sequencing data from seafloor sediments using Oxford Nanopore Technologies.

This study compares Illumina and Nanopore 16S rRNA sequencing for predicting seafloor ecological state using machine learning and feature selection, showing comparable performance with Nanopore as a viable alternative.

The paper introduces linc2function, a deep learning model that identifies and functionally annotates long noncoding RNAs using sequence, structure, and interactome features across species.