IGV: Load custom data (chr, start, end, strand, sample)

Question

[Nicolas Rosewick]

Hi,

I've some data that I want to load into IGV. It represents some viral integration sites for a bunch of samples. For each integration site I've :

• chromosome
• start
• end (in fact start+1)
• strand (tell us the orientation of the virus relative to the reference genome)
• sample id

What I want is to have a separate track for each sample id and to have different colors for + and - strand integration sites ? Is that possible ?

Thanks

Answer 1

[dariober]

To have each sample on a separate track you need a file per sample. To visualize features depending on strand, you could make your tracks as bedGraph (format: chrom, start, end, score). In the score field put 1 if strand is positive, -1 otherwise. Then in IGV you can assign different colour to values positive or negative.

To split your samples in individual files you can first sort by position (sort -k1,1 -k2,2n samples.txt > sorted.txt) then put one sample per bedgraph file with score as above with something like:

prev=''
while read line
do
    sample=`echo $line | cut -f 5 -d ' '` ## Get sample id
    if [[ $sample != $prev ]]
    then
        bedgraph=${sample}.bedGraph ## Initialize new file name
    fi
    ## Output bedGraph line
    echo $line | awk -v OFS='\t' '{if($4 == "+"){strand= 1}else{strand= -1}  print $1, $2, $3, strand}' >> $bedgraph 
done < sorted.txt

(I assume you don't have hundreds of samples otherwise it would be messy)

[Nicolas Rosewick]

Yes that was one of my first idea but I would like to avoid multiple files.