A Python-based optimization framework for high-performance genomics

• Published in: bioRxiv
• Main Authors: Shajii, Ariya, Ibrahim Numanagić, Leighton, Alexander T, Haley Greenyer, Amarasinghe, Saman, Berger, Bonnie
• Format: Paper
• Language: English
• Published: Cold Spring Harbor Cold Spring Harbor Laboratory Press 30.10.2020; Cold Spring Harbor Laboratory
• Edition: 1.1
• Subjects: Bioinformatics; Computer applications; Computer programs; Genomics; Haplotypes; Next-generation sequencing; Optimization techniques; Software; high-performance; domain-specific languages; Computational genomics; sequencing
• ISSN: 2692-8205; 2692-8205
• DOI: 10.1101/2020.10.29.361402

Abstract Exponentially-growing next-generation sequencing data requires high-performance tools and algorithms. Nevertheless, the implementation of high-performance computational genomics software is inaccessible to many scientists because it requires extensive knowledge of low-level software optimization techniques, forcing scientists to resort to high-level software alternatives that are less efficient. Here, we introduce Seq—a Python-based optimization framework that combines the power and usability of high-level languages like Python with the performance of low-level languages like C or C++. Seq allows for shorter, simpler code, is readily usable by a novice programmer, and obtains significant performance improvements over existing languages and frameworks. We showcase and evaluate Seq by implementing seven standard, widely-used applications from all stages of the genomics analysis pipeline, including genome index construction, finding maximal exact matches, long-read alignment and haplotype phasing, and demonstrate its implementations are up to an order of magnitude faster than existing hand-optimized implementations, with just a fraction of the code. By enabling researchers of all backgrounds to easily implement high-performance analysis tools, Seq further opens the door to the democratization and scalability of computational genomics. Competing Interest Statement The authors have declared no competing interest. Footnotes * ↵1 Lead Contact * https://github.com/seq-lang/seq * https://seq-lang.org