Some transcripts sequences have no prediction output

[creaturemoon]

Hi,
Thank you for developing this tool. I found that some transcripts did not appear in the output files when using CodAn to predict CDSs of novel transcripts in fruit flies with Invertebrate Full and Partial models, while other transcripts have expected results. I’m wondering why the transcripts haven’t been output. Is there a way I can get the normal results?

Here is a transcript sequence that has no output.

TCGCGCCCTAAACTTCTGACCGCACAACACTTCGACAGCTCAAACAGTGCGCATTAACTTTTTGCCTTTC
TGTGAAATCGGAATTcccctttatttatttTGAcaacGTACCCCTACAGATTCTTCTCCCAATGTCCTGG
GCCAATGTCCATCCATGGCCACTTGTCCGAGCTGCGGCGTCCGCAGGGAGACGAAGGTGGAGTTCGAGCC
CAACACCAAGACCCACCTGATGGCCCTCCTCATCTGCATGCTGGGgggcatttgctgctgctgcattccG
TACTGCACCGACTCCTGCCAGTCGGCCAAACAcacctgcagcagctgcggaGCCTTCGTGGGCACCTACA
AGAACTAGGAGTGCGGTCTCAACGCCAAATATTACAGGAATGGACAGATCCCACTTTGTATTTCTTATTG
ATGGTATCAGATTTGGCAcgaatatttacaaaatacaaCTTAATTGAATATCCTTTTCGCATTGGTTTac
tctttaaaataaATGtataatcaCACAAA

Thank you. Looking forward to your reply.

[pedronachtigall]

Hi @creaturemoon ,

Thank you for using CodAn in your analysis.

It is been a while since the development and I am not sure if the soft-masking (lower case letters) can be an issue for the prediction. Can you upper case all letters and re-run it?
A command like the following may serve you well:
awk 'BEGIN{FS=" "}{if(!/>/){print toupper($0)}else{print $1}}' in.fna > out.fna

Another reason can be related to the fact that the prediciton models is not finding any confident CDS in the sequence.
It can be a transcript representing a ncRNA or the model is not sensitive enough for this specific case.

Hope that it helps you and sorry for the delay.

Best,
Pedro