Sreeram Kannan

HQAlign improves structural variant detection from nanopore reads by incorporating sequencing physics and SV-specific alignment adjustments, capturing more SVs than minimap2 with comparable performance.

This paper introduces Scrambler networks, a method for interpreting sequence-based neural networks by learning stochastic masks to identify salient sequence positions and explain model predictions.

CellMeSH automatically annotates cell types from scRNA-seq data using a probabilistic method and a literature-based database of gene-cell type associations.

QAlign improves alignment accuracy and rates for nanopore reads by converting them into current-level representations before alignment.

UNCURL-App provides an interactive tool for scRNA-seq analysis, integrating cell type databases and allowing user-driven re-analysis of clusters and pathways.

This paper introduces Scribe, a toolkit for causal gene regulatory network inference from single-cell data, showing that pseudotime data performs worse than true time series data.

UNCURL improves scRNA-seq preprocessing by incorporating prior knowledge and scaling to large datasets, enhancing clustering and lineage estimation performance.

Shannon is an RNA-Seq assembler that uses information theory to optimally reconstruct transcripts, achieving higher sensitivity and specificity than existing methods.