llrnaseq-rna-features-pipeline

This readme explains how to use the Nextflow llrnaseq in conjunction with the rna-features python package to generate transfer learning expression features.

Process

1. Install Nextflow (>=21.04.3) and the llrnaseq pipeline.
2. Install rna-features (python>=3.9).
3. Create a samplesheet.csv that points to your fastq.gz reads.
4. Run llrnaseq with the appropriate options for the sample species.
5. Perform the appropriate differential expression analysis on the gene counts.txt produced by llrnaseq using DESeq2. A sample R script along with inputs and expected output contrast files can be found in the deseq2_example folder.
6. Repeat steps 3 - 5 on each dataset, storing the tpm.tsv (produced by llrnaseq) and contrast files (produced by DESeq2) in folders by dataset.
7. Generate an expression features matrix by passing the dataset directory paths to rna-features.