During the course, we developed a pipeline for RNAseq analysis, from generating genome indices and quality control, to trimming, aligning and differential expression analysis. This week's homework includes creating additional rules for QC and visualising the results of the expression expression analysis using PCA.
Look at the documentation of reseqc: http://rseqc.sourceforge.net/ Find the usage information on how to figure out the strandedness of your library. Include a rule calling that command in your workflow and visualise via your multiqc rule.
Use the pca.R script in the scripts folder to visualise the expression data in a PCA plot. As input it requires the 'deseq2/all.rds' output from the setup_de rule named as dds. As output it requires the name 'pca_plot'; you can choose the output file name, but it should end in 'pdf' (have a look at the script file if you are interested what's happening under the hood.). The output name in the rule should follow the way we have called the table output in the 'deseq2' rule. Use the deseq2.yaml as conda environment.
Have a look at the output files of 'deseq2' rule. Use the command line to figure out how many up and down-regulated genes there are (no need to include in snakemake).
Snakemake has the neat functionality of generating a final html report. Have a look at this link: https://snakemake.readthedocs.io/en/stable/snakefiles/reporting.html and generate a final html report of your pipeline.
Send your final snakemake file, your html report and the number of differentially expressed genes for grading.