Test ConvertGDb genbank in Windows

[ssnn]

There is some glitch with ConvertDb genbank and the wrapping of tbl2asn. Need to test this function in Windows.

Sometimes tbl2asn fails to recognize the input fasta (see also #137):

There is a known issue with the --asn argument. In some environments, for reasons that are presently unknown, tbl2asn may fail to recongizing the input fasta file and report an error stating Unable to read any FASTA records. Running tbl2asn manually should resolve the issue.

Sometimes some sequences are missing in the output (see also #162):

It is recommended that the number of sequences in the output sqn file be verified against the number of sequences in the input tab or tsv file. From the command line, this can be achieved via grep -c iupacna *.sqn. This step is not necessary if running tbl2asn outside ConvertDb.

‌

[Jason Vander Heiden]

For the missing sequences, is there a discrepancy between the number of records in the output and the PASS/FAIL counts shown in the console log? Filtering of incomplete sequences is intentional.

[Julian Zhou]

The console log says all passed. But when you actually look in the sqn, not all sequences are there. Problem goes away when running tbl2asn by itself outside ConvertDb. Also, since I ran ConvertDb a few times with --asn on as I was doing resubmission multiple times, I eventually noticed that the number of sequences in sqn varied slightly from submission to submission, suggesting that the number of sequences being omitted each time was different.

[Hailong Meng]

Tested and it works well