Pipeline for taking STAR's SJ.out files and parsing the counts for a given bed of named spliced junctions
Edit the top lines of the .smk file to run correctly
There's two independent parts to the analysis, one will turn STAR's splice junction into bed files, sort them, and then use bedtools plus some awk to give a final output file that looks like this(without the header)
| chromosome | start | end | filename_this_count_comes_from | count | strand | name_of_junction_in_your_input |
|---|---|---|---|---|---|---|
| chr19 | 7168094 | 7170537 | Cont-B_S2.SJ.out | 49 | - | INSR_annotated |
| chr19 | 7168094 | 7170537 | Cont-C_S3.SJ.out | 30 | - | INSR_annotated |
| chr19 | 7168094 | 7170537 | Cont-D_S4.SJ.out | 35 | - | INSR_annotated |
| chr19 | 7168094 | 7170537 | control_fluorescent_2.SJ.out | 9 | - | INSR_annotated |
| chr19 | 7168094 | 7170537 | control_fluorescent_3.SJ.out | 5 | - | INSR_annotated |
| chr19 | 7168094 | 7170537 | control_none_1.SJ.out | 20 | - | INSR_annotated |
To use that part properly you'll want to edit parse_star_junctions.smk
And tweak the following input
project_dir - this is a top level folder where the sorted beds, and outputs are going to end up
out_spot - a folder underneath project_dir that will be created, and sorted beds are output is going to appear
bam_spot - pipeline is fairly lazy, it's going to glob wild cards from this folder, so make sure all the samples you want to are in
the same folder, (symlinks are fine!)
bam_suffix - suffix of the bams for pattern matching to work
sj_suffix - suffix of your splice junction tables for pattern matching to work
bed_file - a bed file of junctions you want to compare against
final_output_name - a name for your file. the final output file will be located in:
{project_dir}/{out_spot}/{final_output_name}.aggregated.clean.annotated.bed
This will contain only the junctions in bed_file and with the names of the junction and the names of the file it was found it
You'll also have a file called
{project_dir}/{out_spot}/{final_output_name}.aggregated.bed
This is all the junctions which overlapped the ones in bed_file, useful to check
if you expected junctions that weren't present because you might have an a one-off error.
Basic work flow is that the first rule will convert a SJ.out.tab to a bed file, and put the 'name' of each entry as the name of the file it's in
e.g. if I input a folder with samples called sample01.SJ.out.tab and sample02.SJ.out.tab
I'll get 2 beds that look like this
chrY 57208979 57209532 sample01.SJ.out 0 + chrY 57209059 57209219 sample01.SJ.out 0 +
chrY 57208979 57209532 sample02.SJ.out 0 + chrY 57209059 57209219 sample02.SJ.out 0 +
The second part uses Dasper to annotate relative to a GTF and convert STAR's splice junction counts to percent spliced in.
Feel free to try linking the 2 together but that part is highly developmental yet, so buyer-beware.