Research Workflows - Towards reproducible science via detailed provenance tracking in Open Science Chain

被引：2

作者：

Nandigam, Viswanath ^{[1
]}

Lin, Kai ^{[1
]}

Shantharam, Manu ^{[1
]}

Sakai, Scott ^{[1
]}

Sivagnanam, Subhashini ^{[1
]}

机构：

[1] Univ Calif San Diego, San Diego Supercomp Ctr, La Jolla, CA 92093 USA

来源：

PRACTICE AND EXPERIENCE IN ADVANCED RESEARCH COMPUTING 2020, PEARC 2020 | 2020年

基金：

美国国家科学基金会;

关键词：

Data Reproducibility; Data Provenance; Data Integrity; Blockchain;

D O I：

10.1145/3311790.3399619

中图分类号：

TP31 [计算机软件];

学科分类号：

081202 ; 0835 ;

摘要：

Scientific research has always struggled with problems related to reproducibility caused in part by low data sharing rates and lack of provenance. Credibility of the research hypothesis comes into question when results cannot be replicated. While the growing amount of data and widespread use of computational code in research has been pushing scientific breakthroughs, their references in scientific publications is insufficient from a reproducibility perspective. The NSF funded Open Science Chain (OSC) is a cyberinfrastructure platform built using blockchain technologies that enables researchers to efficiently validate the authenticity of published data, track their provenance and view lineage. It does this by leveraging blockchain technology to securely store metadata and verification information about research data and track changes to that data in an auditable manner. In this poster we introduce the concept of "research workflows", a tool that allows researchers to create a detailed workflow of their scientific experiment, linking specific data and computational code used in their published results in order to enable independent verification of the analysis. OSC research workflows will allow for detailed provenance tracking both within the OSC platform as well as external repositories like Github, thereby enabling transparency and fostering trust in the scientific process.

引用

页码：484 / 486

页数：3