Algorithms and Architectures for Computational Genomics

This introductory lecture outlines the scope of "Algorithms and Architectures for Computational Genomics," setting the stage for future discussions on accelerated genome analysis and design.

This lecture provides foundational knowledge in genome analysis, introducing key concepts and methodologies essential for understanding subsequent topics in computational genomics.

The lecture explores diverse DNA sequencing technologies, detailing their mechanisms from molecular biology principles to the generation of raw genomic data for computational analysis.

This lecture delves into fundamental algorithms and applications for biological sequence search and alignment, crucial for identifying genetic similarities and variations.

This session covers the complex computational problem of genome assembly, examining algorithms that reconstruct complete genomes from fragmented sequencing reads.

This segment introduces GateKeeper, an optimized acceleration technique for genomic sequence pre-alignment filtering, enhancing throughput in DNA sequencing pipelines.

The lecture examines SneakySnake, a specialized accelerator designed to optimize Smith-Waterman sequence alignment, crucial for identifying local similarities in genomic data.

This session introduces GRIM-Filter, an architectural solution for accelerating pre-alignment filtering in genomic analysis, improving efficiency for short read mapping.

This lecture focuses on GenASM, a specialized processing-in-memory architecture designed to accelerate genome sequence assembly algorithms, optimizing computational performance.

This segment explores Scrooge, an efficient accelerator designed for genomic sequence alignment, specifically optimized for approximate string matching operations.

The lecture introduces SeGraM, a hardware accelerator designed to significantly speed up genome-wide k-mer counting, a critical step in many bioinformatics pipelines.

This lecture introduces GenStore, a specialized memory architecture that improves data access patterns and efficiency for genomics workloads, enhancing overall performance.

This session examines MegIS, a novel processing-in-memory system tailored for accelerating large-scale genomic sequence alignment and search operations.

This lecture covers GenPIP, a specialized processing-in-memory architecture designed for efficient genomic sequence processing, particularly for read mapping applications.

This session introduces TargetCall, a highly optimized algorithm and system for variant calling in targeted genomic sequencing data, enhancing diagnostic precision.

This lecture details BLEND, a system for accelerating genome sequencing data analysis by effectively balancing compute and communication, critical for distributed processing.

This session explores AirLift, an efficient data transfer and processing architecture designed to optimize genomic data movement in modern computational systems.