Standardization of CMI-PB Multi-omics Datasets

Updated: 2022.11.25
We are now moving towards the second CMI-PB contest for which we have generated new data. @Joaquin is working on re-harmonizing the data such that we have 3 datasets:

• D0) use the feature overlap between 2020 and 2021 (original and already complete)
• D1) uses the feature overlap between 2020, 2021, and 2022 (under development)
• D2) uses the feature overlap between 2021 and 2022 (under development)

Updated: 2022.03.31
At this point this repository stores the codes to standardize (mostly) the CMI-PB datasets between 2020 and 2021. To access these datasets please check out: https://github.com/joreynajr/cmi-pb-multiomics/tree/main/results/main/cmi_pb_datasets/processed/harmonized

Processing Steps

Download the data by using:

bash workflow/scripts/download_from_webpage_links.sh

Note: Don't use workflow/scripts/download_from_ftp_server.sh, not working due to certificate issues.

Aftewards run the standarizing notebook with Jupyter:

workflow/notebooks/development/Standardize_data.ipynb

Ongoing Problems & Brainstorming

• Normalization Issue for Ab titer data, scaling difference between Abtiter values for 2020 and 2021
  • We can try...