2. Run a test

After you have installed TALLSorts properly, you should run a test to ensure it works.

1. Navigate to the root folder containing TALLSorts. Within this folder, you should see the folder tests

2. Activate the tallsortsenv environment using either

   conda activate tallsortsenv
   

   or

   source activate tallsortsenv
   

3. Run the following command:

   TALLSorts -s tests/test_counts.csv -d test_output
   

   This command calls TALLSorts, takes in the RNA-seq counts matrix tests/test_counts.csv, then outputs the classifier results into the test_output folder, which was created in the directory you are currently in.

   (Note that tests/test_counts.csv is a counts matrix with samples as rows and genes as columns. Gene names are in Ensembl format (eg. "ENSG00000000003"). We will add support for gene symbol and Entrez IDs in a later update.

4. Navigate to test_output and view the following results:

   • probabilities.csv: a table containing the subtype classification probabilities for each sample
   • predictions.csv: a list reporting the most likely subtype for each sample
   • multi_calls.csv: for samples that have multiple predicted subtypes, this csv reports on all those extra predictions
   • prob_scatters.html: essentially an interactive plot of probabilities.csv. A frozen image is given by prob_scatters.png.
   • waterfalls.html: an interactive plot of predictions.csv which also shows the probability of predicted subtypes.. A frozen image is given by waterfalls.png.

References

RNA-seq counts used in tests/test_counts.csv were generated from publicly-available RNA-seq reads published by:

Autry RJ, Paugh SW, Carter R, Shi L, Liu J, Ferguson DC, et al. Integrative genomic analyses reveal mechanisms of glucocorticoid resistance in acute lymphoblastic leukemia. Nature Cancer. 2020;1(3):329-44. | Link to paper | Link to data here and here

Sample names were changed to random strings to preserve anonymity.