1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
|
Title:
BioNix: functional, reproducible bioinformatics workflows
Abstract:
A challenge for computational biologists is to make our analyses
reproducible - that is, easy to rerun, combine, and share, with the
assurance that equivalent runs will generate identical results. Current
best practice aims at this using a combination of package managers,
workflow engines, and containers.
In this talk I will present BioNix, a lightweight library built on the Nix
deployment system. BioNix manages software dependencies, computational
environments, and workflow stages together using a single abstraction:
pure functions. BioNix lets users specify workflows in a clean, uniform
way, with strong reproducibility guarantees.
I will also discuss the application of BioNix to the Stafford Fox Rare
Cancer project. Within this project BioNix has been used to manage a
complicated workflow across >100 whole genome and whole exome sequences. I
will discuss lessons learnt from this application and the future direction
of BioNix.
Bio:
Dr Justin Bedő is the Stafford Fox Centenary Fellow in Bioinformatics
and Computational Biology for Rare Cancers at the WEHI. He was awarded
a PhD by the ANU in 2009 in Machine Learning. His PhD investigated novel
machine learning algorithms to solve bioinformatics problems arising
in plant breeding and cancer genomics. After his PhD he obtained
Postdoctoral experience at NICTA and IBISC (Genopole, Paris) as well as
industrial research experience at IBM Research Australia. His interests
span machine learning, diagnostic genomics, and bioinformatics.
|
