handle KeyError: transcript_id not found

[ygidtu]

1. Genes don't have transcript_id, therefore, just try to get transcript_id from transcript and exon.
2. add try-catch to "transcript_id" to throw a better error message
3. using gzip to handle the gzipped gtf file

Still strong recommend using pysam to handle the IO of bam file

[dgarrimar]

I think this proposal is great, and with it we can solve #2 and #46. I'd just provide a meaningful error message instead of printing the GTF line. I proposed a couple of lines of code, with a structure analogous to other error messages along the script.

[dgarrimar]

Suggested change

	except KeyError as e:
	print(line)
	exit(e)
	except KeyError:
	print("ERROR: 'transcript_id' attribute is missing in GTF file.")
	exit(1)

Maybe better if we do not print the GTF line, but rather provide a more clear error message, following this structure (analogous to the one employed for other error messages along the code).

[dgarrimar]

I take advantage to suggest code to fix #48

[dgarrimar]

Suggested change

	transcript_id = d["transcript_id"]
	transcript_id = re.findall('transcript_id ("[^"]+")', tags)[0]

As it affects the same portion of code I take advantage to suggest this line too, which fixes #48

[dgarrimar]

Hi @ygidtu, the modified proposals have been added, and will be part of the next release. Thank you!