UC 11. Sex as a Biological Variable (Wilson)

[NoopDog]

Interop Contact: Brian O'Connor
Active in 2021: Active
Studies: any WGS from AnVIL, BDCat, CRDC, and GMKF

Title: Sex as a Biological Variable.

Researcher: Melissa A. Wilson (Arizona State University)

[jackDiGi]

@briandoconnor has summarized in a great one-pager, linked to the NCPI Sys Interop charter. Remaining detail to figure out is compute cost - we can use the first few samples processed to help benchmark costs and decide if optimization is necessary.

[briandoconnor]

See this document for a nice writeup by Melissa and some more details: https://docs.google.com/document/d/1y7Mt0JFA4REp2IELlWAVwiIrq_6i7_n3tQOpoxzMeWI/edit#

[jackDiGi]

pinged for the Fall workshop

[linikujp]

Brain developed the demo material here: https://github.com/briandoconnor/ncpi-interop-demo-oct-2021

[linikujp]

Per 11/5/21 Sys. Interop, meeting, this project has 4 phases. Brian's demo above is the phase 1 - proof of concept.
Phase 2: to pull data to Seven Bridges' workspace.
Phase 3: complete the pipeline implementaion in both Tera and SB and apply to all datasets. Melissa needs to apply grant to support this step.
Phase 4: compare the function equivalence in both platforms.
@briandoconnor Please comment.

[NoopDog]

@briandoconnor @jackDiGi this use case is listed on the NCPI staging site here:
https://staging.anvilproject.org/ncpi/demonstration-projects/sex-as-a-biological-variable-wilson

This links out also to Brian's repo for reproducing some of the initial steps of Wilson's work.

Any feedback you have is appreciated.

Cheers,
Dave

[briandoconnor]

Sorry for the late response but this is great!!

Giving an update on this use case today in NCPI Sys Interop. See https://docs.google.com/presentation/d/1ZAKQkl--n_yjJckALBFmG0y6krqFl3p63qaU9MeOedU/edit#slide=id.g11da010df2d_0_9

[linikujp]

Updates on March 25 and June 6, Melissa Wilson has received cloud credits from NHGRI’s AnVIL to scale up her trial analysis.
This will allow us to 1) port her XYalign pipeline to Terra and 2) scale up the testing across AnVIL, BDCat, CRDC, and Kids First data.
Brian O’Connor meet with Brendan Pinto, the postdoc in charge of the effort, and trained on using Terra (early March)
Next immediate steps are 1) ensuring billing is setup in Terra and 2) applying for dbGaP access for chosen datasets.
-updates by March 25.
Currently, Brendan is testing out small data sets (June 6).
Goal is to have the scaled up trial to be completed by Aug 31, 2022 (credits end date)

[DrPintoThe2nd]

Follow-up: we scaled up our analysis to a total of 50 samples for our final project and have submitted this work for publication in Feb. 2023, https://doi.org/10.1101/2023.02.22.529542