Sanchit Misra

This paper presents an optimized AlphaFold2 implementation for CPUs, achieving significant speedups and higher throughput compared to GPU-based solutions.

This work optimizes whole-genome alignment for multi-core processors, improving speed by 1.6x to 7.2x across human, plant, and primate genomes without sacrificing accuracy.

This paper introduces efficient algorithms for polygenic score inference using large-scale genetic data, achieving significant improvements in speed, memory use, and cross-population transferability.

The paper accelerates deep learning-based ATAC-seq analysis on CPUs, achieving 2.27x speedup over GPUs using efficient layers and reduced precision training.

The paper optimizes minimap2 for faster long-read mapping on CPUs, achieving up to 1.8x speedup without changing output.

The paper introduces LISA, a machine learning-based index for accelerating sequence search in genomics, showing performance improvements over traditional indexes in various case studies.

The paper introduces a parallel algorithm for sequence-to-graph alignment, achieving significant speedup for aligning DNA reads to variation graphs.

The paper identifies four computational building blocks that dominate runtime in NGS secondary analysis tools across sequence mapping, de novo assembly, and variant calling.