Negative tRNA coordinate values

[ohickl]

Hi,

I am occasionally encountering negative coordinate values like this:

Sequence                	       	tRNA   	Bounds 	tRNA  	Anti  	Intron Bounds 	Inf   	
Name                    	tRNA # 	Begin  	End    	Type  	Codon 	Begin         	End   	Score  	Note
--------                	------ 	-----  	------ 	----  	----- 	-----         	----  	------ 	------
...
C0000118171  	2      	66     	-2     	Asn   	GTT   	0             	0     	54.5   	
...

Is this intended, and if so, what does it mean?

I will set these to 1 for now since it breaks downstream analyses. Does anything speak against doing this?

Best

Oskar

[patriciaplchan]

This is not intentional. Can you send an example input sequence and the options you use to run tRNAscan-SE so that I can check it out?

[ohickl]

Sure.
I am running tRNAscan-SE through Bakta, like this:

    cmd = [
        'tRNAscan-SE',
        '-G',
        '--output', str(txt_output_path),
        '--fasta', str(fasta_output_path),
        '--thread', "1",
        str(chunk_path)
    ]

Version:

trnascan-se                 2.0.12        pl5321h031d066_0     bioconda

The sequence:

>C0000118171
CCCAACTGGGCTCGAACCAGTGACATCATGATTAACAGTCATGCGCTCTACCGACTGAGCTATGGAGGAATATATGAACCCTGCATCAAAAAGATACAGGGTGATGTCGGCATTACCTATTTTTCCGGGAGGCTGCCCTCCCATTCTACGCCGTCCGTTCCGCGAATAACCACTATACAGTGATAACCTTAATTTTTAATATATTTGTTTGCAATTGCAAAGCTCACATTTTCATTGTTGTCATCATTAAAATTATATGCTTCAATATCACGATAGTTGTATTTTTCAAGTGCTCTTTTAAAACTTTTTGATATACCTGTTCTTAAATTTTCATATTTCGTTTTATCAGCTTGGCGATCGCTTGGCGATACACTTTTTGATTTGCTATATATTGCATCTGCAAGGTCACGGGTTTCAGTATAAAGCTCGTTTGAATATTTCATTCTCGCAATCTTTTCGGATTTTCCTATTGCTCCCGCTGCACCGACAAATTTACAGTCTAATCCCAAAATCCCCTCATTATCGCAAAATCCGTACAACATCCCTGCGTCATCGCTTATGTCAAGGTTGTAATTACCGCCGTTTTTCCAATTAACTTTAAACGAAATATTGTATTCGTAATCTGATGAACTTGCCCAAACTCTATCTTTTTGGATTGTACCTGAAATTTCTTCGGAATACGACGTATTACCGTCAACAAAAAATGTAGCAGTAAAACTTTGATTATCATAATTTATTGTTTTTATTACTAAGTGAATAGTATAGCCGCTTTTCTTGCATGTATATTCGCAACCAATAACTCTGAATTCACCTGTTGCGTCACCCATAGCATAGCTTGCTGTAGGTATAAATGTAACCAATGATGTTAAAATTAAAATAAAACAAATAAAATTTTTGATTTTTTTCATGATTTCACCTTTAAGTTGTAATTTTTCACAAACATTTTAATATTTCTATAAGCTAAAACCTATTAACAATATCAATAAATGTATAAACAAAAAACATAGCAATTGGTTGATAAGTCACTCTTAAAATCATAAAAAACAACAATGTTATATAATTGATTTTATTAAAATATAGCAAAACAATATTTATAACTACGTATATAAGCGGGAAAAAAGTATAATAAATCCAGTACAAATCCAACGGATTATCCAACAAAAAATTCCACGCGTAAGCGCCTTCATCACAGATTACATAAATCATTTCAGATAAAAATAATACAGTGGATATTATAAATATTATTTTTTTTATTTTACTTAATTGGTTAAAATTCATAATTACCCTTTTTGTTACAATCATTATTTCCGCTCCTAACAATACACCATTATTATAAATTATCAATCTATGATTCCTAACTTAAGGTGATTTCTTCCAATTATATTTATAACATATTTACATTTTGCTTGCAAGAGCATATATATTATATAATATAATGCCTAATTTTAAGAACGTGAACTTCTCCTTTATCGAAAGAATAATAATAAAAAAATATAATATAAATAAGTCCGAAAAAAACACGCCGAAATACGACTTACTAATTCTTTACATAATGACGTTTAAAATATTGCAGATTTTTGACTTGTAAAAAAAATTTGAAAAAGAAAAATCCCATATCAAACAGGTGGTTTCTTGTATCCTTTTCAAGCAAACCTGCTAAAGACGTAAAAAAAAGCAAGTCAATTAAGACTTGCTTTTCGTGGCTCCCCCAACTGGGCTCGAACCAGTGACATCATGATTAACAGTCATGCGCTCTACCGACTGAGCTATGGAGGAATATATGAACCCTGCATCAAAAAGATACAGGGT

[patriciaplchan]

Thanks for sending the sequence! It looks like a bug that occurs when the predicted gene has an expectedly short acceptor stem or is potentially truncated at the end of the input sequence. I will work on including a fix in the next release.