dammit_annotations_parsing

example_gff3_dammit_output

##gff-version 3.2.1
Transcript_0	transdecoder	CDS	1664	2062	.	+	.	ID=cds.Transcript_0|m.1;Parent=Transcript_0|m.1
Transcript_0	transdecoder	exon	1	2064	.	+	.	ID=Transcript_0|m.1.exon1;Parent=Transcript_0|m.1
Transcript_0	transdecoder	five_prime_UTR	1	1663	.	+	.	ID=Transcript_0|m.1.utr5p1;Parent=Transcript_0|m.1
Transcript_0	transdecoder	gene	1	2064	.	+	.	ID=Transcript_0|g.1;Name=ORF%20Transcript_0%7Cg.1%20Transcript_0%7Cm.1%20type%3A3prime_partial%20len%3A134%20%28%2B%29
Transcript_0	transdecoder	mRNA	1	2064	.	+	.	ID=Transcript_0|m.1;Parent=Transcript_0|g.1;Name=ORF%20Transcript_0%7Cg.1%20Transcript_0%7Cm.1%20type%3A3prime_partial%20len%3A134%20%28%2B%29
Transcript_0	transdecoder	three_prime_UTR	2063	2064	.	+	.	ID=Transcript_0|m.1.utr3p1;Parent=Transcript_0|m.1
Transcript_100002	transdecoder	CDS	1	1047	.	+	.	ID=cds.Transcript_100002|m.114344;Parent=Transcript_100002|m.114344
Transcript_100002	transdecoder	CDS	1411	1722	.	+	.	ID=cds.Transcript_100002|m.114346;Parent=Transcript_100002|m.114346
Transcript_100002	transdecoder	CDS	936	1316	.	-	.	ID=cds.Transcript_100002|m.114345;Parent=Transcript_100002|m.114345
Transcript_100002	transdecoder	exon	1	1767	.	+	.	ID=Transcript_100002|m.114344.exon1;Parent=Transcript_100002|m.114344
Transcript_100002	transdecoder	exon	1	1767	.	-	.	ID=Transcript_100002|m.114345.exon1;Parent=Transcript_100002|m.114345
Transcript_100002	transdecoder	exon	1	1767	.	+	.	ID=Transcript_100002|m.114346.exon1;Parent=Transcript_100002|m.114346
Transcript_100002	transdecoder	five_prime_UTR	1	1410	.	+	.	ID=Transcript_100002|m.114346.utr5p1;Parent=Transcript_100002|m.114346
Transcript_100002	transdecoder	five_prime_UTR	1317	1767	.	-	.	ID=Transcript_100002|m.114345.utr5p1;Parent=Transcript_100002|m.114345
Transcript_100002	transdecoder	gene	1	1767	.	+	.	ID=Transcript_100002|g.114344;Name=ORF%20Transcript_100002%7Cg.114344%20Transcript_100002%7Cm.114344%20type%3A5prime_partial%20len%3A349%20%28%2B%29
Transcript_100002	transdecoder	gene	1	1767	.	-	.	ID=Transcript_100002|g.114345;Name=ORF%20Transcript_100002%7Cg.114345%20Transcript_100002%7Cm.114345%20type%3Acomplete%20len%3A127%20%28-%29
Transcript_100002	transdecoder	gene	1	1767	.	+	.	ID=Transcript_100002|g.114346;Name=ORF%20Transcript_100002%7Cg.114346%20Transcript_100002%7Cm.114346%20type%3Acomplete%20len%3A104%20%28%2B%29
Transcript_100002	transdecoder	mRNA	1	1767	.	+	.	ID=Transcript_100002|m.114344;Parent=Transcript_100002|g.114344;Name=ORF%20Transcript_100002%7Cg.114344%20Transcript_100002%7Cm.114344%20type%3A5prime_partial%20len%3A349%20%28%2B%29
Transcript_100002	transdecoder	mRNA	1	1767	.	-	.	ID=Transcript_100002|m.114345;Parent=Transcript_100002|g.114345;Name=ORF%20Transcript_100002%7Cg.114345%20Transcript_100002%7Cm.114345%20type%3Acomplete%20len%3A127%20%28-%29
Transcript_100002	transdecoder	mRNA	1	1767	.	+	.	ID=Transcript_100002|m.114346;Parent=Transcript_100002|g.114346;Name=ORF%20Transcript_100002%7Cg.114346%20Transcript_100002%7Cm.114346%20type%3Acomplete%20len%3A104%20%28%2B%29
Transcript_100002	transdecoder	three_prime_UTR	1048	1767	.	+	.	ID=Transcript_100002|m.114344.utr3p1;Parent=Transcript_100002|m.114344
Transcript_100002	transdecoder	three_prime_UTR	1723	1767	.	+	.	ID=Transcript_100002|m.114346.utr3p1;Parent=Transcript_100002|m.114346
Transcript_100002	transdecoder	three_prime_UTR	1	935	.	-	.	ID=Transcript_100002|m.114345.utr3p1;Parent=Transcript_100002|m.114345
Transcript_100003	shmlast.LAST	conditional_reciprocal_best_LAST	48	538	8.600000e-184	+	.	ID=homology:230094;Name=gi|768937493|ref|XP_011608972.1| PREDICTED: uncharacterized protein LOC105417370 [Takifugu rubripes];Target=gi|768937493|ref|XP_011608972.1| PREDICTED: uncharacterized protein LOC105417370 [Takifugu rubripes] 5 498 +;database=protein.fa

only_matched_referece

# loops through directories with annotations from multiple species
# sorts each by ID
# drops all rows not annotated to cutsom reference protein database, startswith("gi")
# returns transcript ID and gene name
# writes to .csv file

import pandas as pd
import os 
# requires dammit env:
# source activate py3.dammit
from dammit.fileio.gff3 import GFF3Parser

dammit_dir = '/home/ljcohen/osmotic_damit/'
dammit_dirs = os.listdir(dammit_dir)
print(dammit_dirs)
for dammit_dirname in dammit_dirs:
    if dammit_dirname != "sbatch_files":
        genus_species = dammit_dirname.split(".")[0]
        dammit_gff = dammit_dir + dammit_dirname + "/" + genus_species + ".trinity_out.Trinity.fasta.dammit.gff3"
        print(dammit_gff)
        annotations = GFF3Parser(filename=dammit_gff).read()
        all_names = annotations.sort_values(by=['seqid'],ascending=True)[['seqid','Name']]
        annotations = annotations.dropna(subset=['Name'])
        fund = annotations[annotations['Name'].str.startswith("gi")]
        names = fund.sort_values(by=['seqid'], ascending=True)[['seqid', 'Name']]
        names_out = '/home/ljcohen/osmotic_assemblies_farm/'+genus_species+'.trinity_out.Trinity.fasta.Fundulus.genenames.csv'
        #names.to_csv(names_out)
        #print("Written:",names_out)
        all_names_out = '/home/ljcohen/osmotic_assemblies_farm/'+genus_species+'.trinity_out.Trinity.fasta.all_gene_names.csv'
        all_names.to_csv(all_names_out)
        print("Written:",all_names_out)

sortby_evalue

# takes Trinity .fasta file
# sorts by ID and e-value
# checks that score < 1e-05
# drops duplicates (picks lowest e-value)
# returns transcript ID and gene name
# writes to .csv file

import pandas as pd
from dammit.fileio.gff3 import GFF3Parser

annotations = GFF3Parser(filename='mahi.trinity_out.Trinity.fasta.dammit.gff3').read()
alignments = annotations.query('source != "transdecoder"')
names = alignments.sort_values(by=['seqid', 'score'], ascending=True).query('score < 1e-05').drop_duplicates(subset='seqid')[['seqid', 'Name']]
names.to_csv('genenames.csv')