InfectionMedicineProteomics / BINN

Generating biologically informed neural networks
https://infectionmedicineproteomics.github.io/BINN/
MIT License
25 stars 10 forks source link

Input clarification #27

Closed danielgeiszler closed 1 year ago

danielgeiszler commented 1 year ago

Can you please clarify what the input format is? Is the minimum input format protein names and a quant matrix? The example input is only a "Protein" column, but the test_data.csv has quite a few other columns.

ErikHartman commented 1 year ago

Hey! The minimum input format to create a network is:

  1. A data file. This really just needs the input column, in our case "Protein".
  2. The network-file.
  3. Alternatively, a translation between input and network, which in our case was UniProt IDs and Reactome IDs.

test_data.csv should not have been in the example. We have now replaced it with test_qm.csv which contains what you'd expect. Hopefully this clears it up.

We also added this image to the README for clarity:

data_explanation

danielgeiszler commented 1 year ago

Thanks! That clears it up immensely.