The ENCODE RNA-SEQ-PIPELINE includes a QC step that counts reads by gene type. This step requires parsing transcript_id <-> gene_type mapping. Purpose of this repo is to document the process of creating that mapping. The code will be included as a jupyter notebook. The annotation files will be included in the repo in folder annotations_and_spikeins. Separate mappings as .tsv files are in folder mappings. The mappings will be created for human annotation version hg19, human annotation v24 and mouse annotation version mm10 a.k.a GRCm38.p4 M7 matching tRNAs downloaded from gencode will be included. Additionally, spikein files from ENCFF001RTP and from ENCFF335FFV will be used. For hg19 and mm10 the merging is done using the merge_annotation.py python script with following command lines run in the annotations_and_spikeins folder using python 3.4.3:

min-tb:annotations_and_spikeins otto$ python3 ../merge_annotation.py --annotation gencode.v19.annotation.gtf.gz --tRNA gencode.v19.tRNAs.gtf.gz --spikeins ENCFF001RTP_spikes_ENCFF335FFV_spikes.fasta.gz --output_filename gencode.v19.trna.ercc.phix.gtf.gz

min-tb:annotations_and_spikeins otto$ python3 ../merge_annotation.py --annotation gencode.vM7.annotation.gtf.gz --tRNA gencode.vM7.tRNAs.gtf.gz --spikeins ENCFF001RTP_spikes_ENCFF335FFV_spikes.fasta.gz --output_filename gencode.vM7.trna.ercc.phix.gtf.gz

Subsequently the transcript id and gene type are extracted from the file resulting from the merge_annotation.py step by running the accompanied python script build_transcript_id_to_gene_type_tsv.py. For human GRCh38 the file where annotation, tRNAs and spikeins are contained is downloaded from the ENCODE portal and used directly in this step. Example command line looks like this:

min-tb:github otto$ python3 build_transcript_id_to_gene_type_tsv.py --input_annotation annotations_and_spikeins/gencode.vM7.trna.ercc.phix.gtf.gz --output_tsv gencode.vM7.trna.ercc.phix.transcript_id_to_genes.tsv

rRNA and Mt_rRNA removal

Within annotations_and_spikeins folder are also annotation files (with spikeins), from which all rRNA annotations are removed, and files from which rRNA, excluding Mt_rRNA annotations are removed. This was done by using unix shell commands in the following way:

Remove all rRNA.

As an example, the shell command to remove all rRNA annotations from ENCFF824ZKD_gencodeV24pri-tRNAs-ERCC-phiX.gtf.gz

min-tb:annotations_and_spikeins otto$ gzip -cd ENCFF824ZKD_gencodeV24pri-tRNAs-ERCC-phiX.gtf.gz | grep -v rRNA | gzip -n > ENCFF824ZKD_gencodeV24pri-tRNAs-ERCC-phiX-minus-rRNA-and-Mt-rRNA.gtf.gz

Remove rRNA , leave Mt_rRNA in.

As an example, the shell command to remove rRNA annotations, leaving Mt_rRNA annotations untouched from gencode.vM7.trna.ercc.phix.gtf.gz.

min-tb:annotations_and_spikeins otto$ gzip -cd gencode.vM7.trna.ercc.phix.gtf.gz | grep -v \"rRNA | gzip -n > gencode.vM7.trna.ercc.phix.minus.rRNA.gtf.gz