Re-opening open-source science through AI assisted development

TL;DR

This paper demonstrates how AI can assist a single human in modifying large open-source scientific software, exemplified by STAR-Flex, to enhance accessibility and re-open science for community review.

Contribution

It introduces a novel AI-assisted development approach that enables rapid modification of complex open-source scientific codebases, exemplified by the creation of STAR-Flex.

Findings

AI enabled the addition of 16,000 lines of code to STAR

STAR-Flex is the first open-source software for Flex data processing

The approach facilitates community review and vetting of AI-generated code

Abstract

Open-source scientific software is effectively closed to modification by its complexity. With recent advances in technology, an agentic AI team led by a single human can now rapidly and robustly modify large codebases and re-open science to the community which can review and vet the AI generated code. We demonstrate this with a case study, STAR-Flex, which is an open source fork of STAR, adding 16,000 lines of C++ code to add the ability to process 10x Flex data, while maintaining full original function. This is the first open-source processing software for Flex data and was written as part of the NIH funded MorPHiC consortium.