igvxml.py

# parse args
import argparse
import subprocess

def escape(s):
    s = s.replace("<", "&lt;")
    s = s.replace(">", "&gt;")
    # this has to be last:
    s = s.replace("&", "&amp;")
    return s

def translate_to_s3(f):
	cmd = "bs -c RADY link file -i " + str(f) +" | xargs curl -w %\{redirect_url\}"
	return escape(subprocess.check_output(cmd, shell=True).decode())

def parse_args():
	parser = argparse.ArgumentParser(description = "write a minimal IGV xml that allows streaming BAM/BAI")

	# reference
	parser.add_argument("-ref", required = True,  default='hg38', choices=['hg19', 'hg38'] )

	# input files	
	parser.add_argument("-bams", required = True, help = "comma-deliminated bam BSSH file ids")
	
	# output xml
	parser.add_argument("-o", required = True, help = "output file path")

	args = parser.parse_args()

	# check args: if indices and names have same length with paths
	return args

args = parse_args()
filelist = [{'bai': translate_to_s3(int(bam)+1), 'bam':translate_to_s3(bam)} for bam in args.bams.split(',')]

with open(args.o, 'w') as xml:
	# headers
	xml.write("<?xml version=\"1.0\" encoding=\"UTF-8\" standalone=\"no\"?>\n")
	xml.write("<Session genome=\""+args.ref+"\" hasGeneTrack=\"true\" hasSequenceTrack=\"true\" locus=\"All\" nextAutoscaleGroup=\"4\" path=\"\" version=\"8\">\n")
	xml.write("\t<Resources>\n")

	# file handlers
	for f in filelist:
		xml.write("\t\t<Resource index=\"{bai}\" path=\"{bam}\"/>\n".format(bai = f['bai'], bam = f['bam']) )
	
	# end
	xml.write("\t</Resources>\n")
	xml.write("</Session>")