BCC2020

→ Abstract


The presenter(s) will be available for live Q&A in this session (BCC West).

Andrea Guarracino 5, Peter Amstutz 2, Thomas Liener 3, Michael Crusoe 4, Adam Novak 6, Erik Garrison 6, Tazro Ohta 7, Bonface Munyoki 1, Danielle Welter 8, Sarah Zaranek 2, Alexander (Sasha) Wait Zaranek 2, Pjotr Prins 1

1 Department of Genetics, Genomics and Informatics, The University of Tennessee Health Science
Center, Memphis, TN, USA.
2 Curii Corporation, Boston, MA, USA.
3 independent.
4 Department of Computer Science, Faculty of Sciences, Vrije Universiteit Amsterdam, The
Netherlands.
5 Centre for Molecular Bioinformatics, Department of Biology, University Of Rome Tor Vergata,
Rome, Italy.
6 UC Santa Cruz Genomics Institute, University of California, Santa Cruz, CA, USA.
7 Database Center for Life Sciences, Tokyo, Japan.
8 Luxembourg Centre for Systems Biomedicine, University of Luxembourg, Luxembourg.

Project Website: http://covid19.genenetwork.org/
Source Code: https://github.com/arvados/bh20-seq-resource
License: Apache 2.0

As part of the COVID-19 Virtual Biohackathon 2020 we formed a working group to create a
COVID-19 Public Sequence Resource (COVID-19 PubSeq) for SARS-CoV-2 virus sequences. Our goal
was to create a repository that had a low barrier to entry for uploading and analyzing sequence
data. We followed FAIR data practices: data are published with public domain (CC0) or creative
commons 4.0 (CC-BY-4.0) license, structured metadata is validated against standard ontologies,
and, most importantly, reproducible workflows are executed after the upload in order to provide
up-to-date results rapidly and in standardized data formats.

Existing data repositories for viral data include GISAID, EBI ENA and NCBI. These repositories allow
for free sharing data, but do not enforce strict quality control on submitted data or metadata, and
do not add value in terms of running additional analysis. In addition, some databases have a
restricted license which prevents data from being used in online web services and on-the-fly
computation, hindering research.

We created a prototype sequence resource within one week by leveraging existing technologies,
such as the Arvados Cloud platform (http://arvados.org), Common Workflow Language (CWL)
(http://commonwl.org), and the many free and open source software packages that are available
for bioinformatics. Pipelines developed by several teams were combined into an omnibus
pangenome analysis workflow. Computing resources for this project were generously donated by
Amazon Web Services.