Add the denominator detection in awk

[Wang-Taiwei]

Hello,

When I ran the pipeline I met an error aborting the program and here is the log:

awk 'BEGIN{FS=OFS=" "} FNR>0 && FNR==NR{a[$4]=$4 OFS $0; next} FNR>0{print $0,a[$4]?a[$4]:" "}' /Path/to/TSS_shared_density.bed /Path/to/gene_body_shared_density.bed | awk -v OFS=' ' '{ if ($6 == "+"){print $9, $10, $3, $4,sqrt((($15+$7)/sqrt(($3-$10)^2))^2),($15/sqrt(($11-$10)^2))/($7/sqrt(($3-$2)^2)), $6} else {print $9, $10, $3, $12,sqrt((($15+$7)/sqrt(($10-$2)^2))^2),($15/sqrt(($11-$10)^2))/($7/sqrt(($3-$2)^2)), $6}}' | env LC_COLLATE=C sort -k1,1 -k2,2n > /Path/to/tmpfile (1056631,1056632,1056633)

awk: cmd. line:1: (FILENAME=- FNR=12754) fatal: division by zero attempted

And I am not sure if this is normal for zero to appear in the denominator, and I check there is only one gene which did that.

In order to enable the next procedure to proceed，I add a denominator detection in the peppro.py line 3267, and don't know if it is appropriate but may be useful for others with the same promblems. Thank you for the nice tool!

cmd5 = ("awk 'BEGIN{FS=OFS=\"\t\"} FNR>0 && " + "FNR==NR{a[$4]=$4 OFS $0; next} " + "FNR>0{print $0,a[$4]?a[$4]:\"\t\"}' " + shared_TSS_density + " " + shared_body_density + " | awk -v OFS='\t' '{ if ($6 == \"+\")" + "{print $9, $10, $3, $4," + "$3!=$10?sqrt((($15+$7)/sqrt(($3-$10)^2))^2):\"NaN\"," + "($11!=$10 && $3!=$2)?($15/sqrt(($11-$10)^2))/($7/sqrt(($3-$2)^2)):\"NaN\", $6} " + "else {print $9, $10, $3, $12," + "$10!=$2?sqrt((($15+$7)/sqrt(($10-$2)^2))^2):\"NaN\"," + "($11!=$10 && $3!=$2)?($15/sqrt(($11-$10)^2))/($7/sqrt(($3-$2)^2)):\"NaN\", $6}}' " + "| env LC_COLLATE=C sort -k1,1 -k2,2n > " + temp.name)