Diploid.sh CPU error

[Isabella Pirozzolo]

Hi all - I am running diploid.sh using 

path/to/juicer/juicer-1.6/CPU/common/diploid.sh -g mm10 -s /shared/vcf/vcf_phased.vcf 

And get the warning:

No input file specified, reading from input stream No such file or directory at /shared/juicer_test/juicer-1.6/scripts/diploid.pl line 79

I saw this was mentioned here: https://groups.google.com/g/3d-genomics/c/sfdynczX5Nc/m/QL6zc-NhBwAJ

But I couldn't find a resolution, just a flagged bug.

I know it refers to this part of the .pl script:

if (scalar(@ARGV)==0) {

  print STDERR "No input file specified, reading from input stream\n";

}

# Global variables for calculating statistics

my %chromosomes;

my %maternal_snps;

my %paternal_snps;

# read in SNP site file and store as multidimensional array

open FILE, $site_file or die $!;

while (<FILE>) {

        my @locs = split;

  my $key = shift(@locs);

        my $ref = \@locs;

        $chromosomes{$key} = $ref;

}

close FILE;

I have modified the diploid.sh script and the command line to include hard paths to all the file locations, am certain the command is running the the directory that contains merged_nodupes.txt, confirmed the vcf is unzipped, and checked that the vcf format is correct and matches the chr syntax used in merged_nodupes. 

Can you please help troubleshoot? I am working on an ec2 instance with cpu no gpu, and juicer 1.6 is installed there with the main pipeline working great.

Thanks so much

-Isabella