search

NIH-NCPI / NCPI_use_case_tracker

This repo is for tracking NCPI interoperability use cases.
3 stars 1 forks source link

UC 1b. Genetic Basis of Congenital Heart Defects (Goldmuntz) #3

Open NoopDog opened 3 years ago

NoopDog commented 3 years ago

Active in 2021: [Active] Researchers: Elizabeth Goldmuntz and Deanne Taylor (CHOP), A.J. Agopian (UTH), Ryan Urbanowicz (UPenn) Platforms: NHLBI BioData Catalyst + Kids First DRC @jackDiGi Platform lead: Jack DiGiovanna

Analysis Question

Congenital heart defects (CHDs) are important birth defects to study due to their high mortality, occurring in about 1% of all live births and 10% of stillbirths. Many individuals born with CHDs need early medical interventions to survive.

In spite of the impact CHDs have on public health, little is known about their etiology. However, CHDs have been shown to have a genetic component, evident by their recurrence risk (~5%) in siblings.

In this research, we intend to study the genetic bases of congenital heart defects using a variant and gene set analysis approaches, machine learning methods, amongst other statistical and genetic analysis models to help fill in the gaps that exist in the understanding of the etiology of CHDs. This will help the scientific community to better understand cardiogenesis and to better assess the risk of disease. Access to this whole-genome sequence data will facilitate our work.

Analysis Plan

  1. Obtain confirmation from appropriate NIH Data Access Committees that these datasets and data uses are allowable and can be used/combined in this manner.
  2. Find and aggregate all relevant CHD data into one workspace (PCGC whole exome, PCGC whole genome, locally sequenced data)
  3. Run Kids First DRC pipelines (currently in CWL on Cavatica) across data that has not already been harmonized to GRChr38
  4. Perform joint genotyping across whole-exome data
  5. Perform joint genotyping across whole-genome data
  6. Aggregate and harmonize phenotypic data for analysis
    1. E.g. severity of defect [Dr. Goldmuntz provided some mild cases.]
  7. Bring own custom methods for gene set analysis and machine learning for downstream analysis
jackDiGi commented 3 years ago

If I understand correctly, this use case has finished successfully. Should we close it @linikujp & @NoopDog ?

linikujp commented 3 years ago

Closed without publication or turning into a training material.

linikujp commented 3 years ago

reopen to track for training material.

linikujp commented 2 years ago

@linikujp and @jackDiGi will check @NoopDog videos and provide help for enrich the video.

linikujp commented 2 years ago

Video posted on staging site. @jackDiGi @linikujp will review. https://staging.anvilproject.org/ncpi/demonstration-projects/genetic-basis-of-congenital-heart-defects-goldmuntz