# Directories * ### containers * The files required to create the two docker containers we used with the Google Genomics API are located here. * If you prefer to just pull the images from DockerHub, use the links below. * TCGA Image (input tar or tar.gz with .fastq or fastq.gz files inside): available [here](https://hub.docker.com/r/pjtatlow/kallisto-image-genomics-tcga/) * CCLE Image (input BAM file): available [here](https://hub.docker.com/r/pjtatlow/kallisto-image-genomics-ccle/) * ### data * Any data required as input to the scripts are located in this directory. * ### matrices * Matrix files created from kallisto output * There are matrices for the estimated counts and for the transcripts per million * CCLE has matrices for the untrimmed FASTQ output, and for the trimmed FASTQ output * TCGA is divided by cancer type to make for smaller, more manageable matrices * The IDs used in the TCGA matrices are CGHubAnalysisIDs, which you can't search on the Genomic Data Commons. If you want to change these IDs to something else, you can use `map_TCGA_id.py` and `TCGA_ID_MAP.csv` to change them to Aliquot Barcodes or Aliquot ID's, just know that neither of these two identifiers are unique. * Ex. `python map_TCGA_id.py path/to/TCGA_ID_MAP.csv path/to/input-matrix.tsv path/to/output-matrix.tsv --aliquot-barcode` * You can also use `--aliquot-id` if you prefer that identifier. * Google BigQuery tables can be found [here](https://bigquery.cloud.google.com/dataset/cgc-05-0002:Cancer_RNA_Seq_recompute). (You must login with your Google Account to access). * ### scripts * Python and R scripts used to submit files, process the output, and create graphs. * See [Script Descriptions](https://osf.io/gqrz9/wiki/Script%20Descriptions/) for details