Skip to content

Latest commit

 

History

History
executable file
·
51 lines (37 loc) · 2.25 KB

iv_bcg_readme.md

File metadata and controls

executable file
·
51 lines (37 loc) · 2.25 KB
Raw

README for datasets relating to IV BCG in NHP

Manuscripts

Hacks related to IV BCG in NHP will build on these studies and the published datasets:

Cleaned datasets and data dictionaries

Bulk transcriptomics

Bulk RNA sequencing from whole blood was published in Liu et al. Data has been aggregated at the gene level, with files specified below. Values are log2-counts.

meta-data

  • filepath: [DATA PATH]/liu_etal_metadata.csv
  • columns: sampleid [OTHERS]
  • rows: one row per animal
  • combines both the Darrah et al. "route" study and "IV dose" studies (2020 and 2023)

counts table (long-form)

  • filepath: [DATA PATH]/liu_etal_counts_long.csv
  • columns: sampleid, gene symbol (macca mulatta), and log2 count
  • rows: one row per sample, per gene
  • can be joined on sampleid with the meta-data CSV
  • combines "route" and "dose" Darrah et al. studies

counts table (wide-form)

  • filepath: [DATA PATH]/liu_etal_counts.csv
  • columns: gene and sampleids (one column per sampleid)
  • rows: one row per gene
  • this format is suitable for reading straight into differential expression analysis, with sampleid order matching that of the sampleid order in the meta-data CSV
  • combines "route" and "dose" Darrah et al. studies

BCG-induced immune responses

Both Darrah et al. studies meausured cellular and humoral immune responses before and after BCG vaccination from lung (BAL) and blood.

immune_responses

  • filepath: [DATA PATH]/darrah_etal_immune_resp.csv
  • columns: sampleid, tissue (blood or BAL), bioid (unique identifier of the immune response), [OTHER COLUMS DESCRIBING IMMUNE RESPONSE], Value, Unit
  • rows: one row per animal
  • combines both the Darrah et al. "route" study and "IV dose" studies (2020 and 2023)
  • importantly, the sampleid match those from the Liu et al. meta-data file above to allow for joining these datasets (e.g., 0NR_65 from "route" study and OBC_wk12Pre from "dose" study)

[KMB: here I'm thinking that the immune response metadata describing which assay etc will be pre-joined withe the assay values themselves. Let me know if you think it should be kept separate]

TODO:

  • Clean and prep each file
  • Write startup scripts
  • Organize repo
  • Test on Rstudio servers