Tag annotations ignored when multiple are present

[xrobin]

The following line from the Ensembl Homo_sapiens.GRCh37.75.gtf is parsed incorrectly:

1   protein_coding  exon    860260  860328  .   +   .   gene_id "ENSG00000187634"; transcript_id "ENST00000420190"; exon_number "1"; gene_name "SAMD11"; gene_source "ensembl_havana"; gene_biotype "protein_coding"; transcript_name "SAMD11-011"; transcript_source "havana"; exon_id "ENSE00001637883"; tag "cds_end_NF"; tag "mRNA_end_NF";

As you can see the line contains two 'tag' attributes. Only the second one is present in the DataFrame returned by read_gtf_as_dataframe:

seqname          source feature   start     end  score strand frame  \
0       1  protein_coding    exon  860260  860328    NaN      +     .

           gene_id    transcript_id exon_number gene_name     gene_source  \
0  ENSG00000187634  ENST00000420190           1    SAMD11  ensembl_havana

     gene_biotype transcript_name transcript_source          exon_id  \
0  protein_coding      SAMD11-011            havana  ENSE00001637883

           tag
0  mRNA_end_NF

The cds_end_NF tag is lost. Ideally both tags should be presented in a list, but I'm not sure if that's possible with pandas.

[iskandr]

Sorry for the very slow response and not sure if this is still relevant to you. I think this PR is trying to address the same issue: #6

In that case, the solution is to concatenate the multiple values in a comma separated string. I think collecting a list makes more sense.