-
Notifications
You must be signed in to change notification settings - Fork 12
3. Under the hood
Keen to find out what you can change? Check out the parameters that can be entered below. If you feel that there is an option that should be implemented, please don't hesitate to contact me.
export CLINKERDIR=/path/to/clinker/root/bpipe -p option1="something" -p option2="something_else" [...] $CLINKERDIR/workflow/clinker.pipe /path/to/*.fastq.gz
bpipe $CLINKERDIR/workflow/clinker.pipe to view the below options
out=/path/to/clinker/output
Where do you want Clinker to put its output
caller=/path/to/fusion/caller/results.csv
The path to the fusion caller output (not required if fusion genes specified)
header=true or header=false
Is there a header row in your fusion caller output. This is a mandatory parameter.
fusions=BCR:ABL1,P2RY8:CRLF2
List of fusions genes/genes of interest to print in GVIZ.
Note: Inclusion of this parameter will print these genes to PDF only if print=true. Note2: Fusion genes are separated by a : in a comma delimited list. i.e. BCR:ABL1,P2RY8:CRLF2
genome=19 or genome=38
Genome version the fusion caller output was created with (Default=19)
del=c or del=t
Delimiter of fusion caller output (c = comma, t = tab) (Default=c)
col=1,2,3,4 or col=1,2
Columns in fusion caller output where genomic coordinates reside. (Default=1,2,3,4)
(chr1,breakpoint1,chr2,breakpoint2) OR (chr1:breakpoint1,chr2:breakpoint2)
Note: This is caller specific, i.e. JAFFA col=3,4,5,6
Note 2: I know... But there are too many different naming conventions for me to risk trying to figure it out. There's a relevant XKCD for this...
competitive=true/false
Include superTranscripts (not fused) in fusion superTranscriptome (Default=true)
Note: This will significantly add time to STAR alignment, but you will get sprurious junctions otherwise.
WARNING This will require more memory to be allocated by genome_mem=X align_mem=Y (approx 32GB).
genome_mem=36000000000
Maximum memory usage for STAR genomeGenerate (bytes)
Make sure this works for your system, otherwise you're machine will crash... amirite?
align_mem=20000000000
Maximum memory usage for STAR alignment (bytes)
threads=1
Maximum threads to use with STAR... Increasing this makes stuff happen faster.
Make sure this works for your system. Can't use more threads than you have... amiriteagain?
This guy knows what I'm talking about...
print=false
Print PDFs of all specified fusion genes in the fusion parameter.
No harm in having this as false, checking out the fusions in IGV and then proceeding.
You won't have to run through STAR again.
pdf_width=16
PDF width (inches)
pdf_height=9
PDF height (inches)
ratio=1,5,1,3,1,3
This one is a bit tricky, but essentially it is the relative size of each track.
For instance, if your track order was axis, coverage, gene track, domain track, transcript track, sashimi plot, then gene track will be 5 times larger than the axis.
support=2
This little beauty allows you to specify the minimum read support you want for splice junctions/fusion junctions. Any junction that has a read support of less than this number will not be represented.
You can change the colours for the final plot, hex code only (no hash)
col_coverage=6e65ad col_gene1=3983AA col_gene2=2b749a col_domains=f05f3b col_transcript=a1d16e col_transcript_bg=f2ffe4 col_transcript_border=215A4A col_junctions=D7D4E4 col_fusions=6e65ad col_highlights=ff6d6d
I recently broke this... Will bring it back shortly after I find out how big of an idiot I am.
No wuckas, just contact me @MCRI or open a support request. Happy to help if I can.