Unable to find ORF for fasta file

[dena.dinesh]

Hi,

I have a list of ids for which I want to extract the corresponding nucleotide sequences from transcritptome.fasta file. I loaded the both my ids and transcriptome into R and later extracted the the nucleotide sequences from transcriptome.fasta file. Later I appended the fasta symbol(>) for all ids and exported it to file.

When I uploaded the file to predict the ORFs, I am unable to predict anything for it. I used the frameDP and getORF from EMBOSS. FrameDP prints a pepdb.fa file without any translated sequences while getORF from EMBOSS throws error saying that all the sequences has zero length. I dont where the problem lies.

Could anyone help me in figuring this out. Kindly help me.

Here is the R code which i used to extract the sequences and print it to file.

library(seqinr)
ids=as.character(read.delim("path/to/ids/file.txt"))
dd=read.fasta("path/to/transcriptome.fasta",seqtype="DNA",as.string=T)
fasta_seq=unlist(dd[names(dd) %in% ids])
names(fasta_seq)=paste(">",names(fasta_seq),sep="")
write.table(fasta_seq,file=paste(dir,"/",name,sep=""))

here the output from which has been generated from R. Initially it had "" from the and later I removed by replaced

"fasta_seq"
">dd_smedV4_1188_0_1" "atttgttccattcataaataaaagtagacggctgaaacagtatataaagctataaaaaattcaaacgtatcactgaaataaaatgatatcatgcagattttgttttcaagtaatctttggattccttttagtattgttccactcagatctagtaatctcgagatattttttgcctccagcagactggacaaattccaatgtttttaacaaaagagacaaacctgcaaacggagtgtcgaatgaattgtcgagagaggtgttcaattgtttacagttttgtgccgaatgtagctatgcgtatggtccgtattttaatgtcttcaaatgcggtagagcgtgtagcagcggtgtcatcaacaacaacaaatccaaggagtgtaagtcaaacataatttaagagagctcgtcgttggagcgagatattttgaggtgtccgcttttcgtgaataaattt"
">dd_smedV4_120_0_1" "tgattgaatggctgcaattatatttcaagtaatttcaattaatatcctaaatgggaaaattagtcagaaaattcgattacattatgaaattcaattattatgagtcctcagtaaaatcatttttattgcccagttatgctataaatacagtcccgacaatcaatattcagtcaaccatgaaattcttaattttagccagtattgcctgtataattctgatgcttactttcgaagcacgatcagatagtccaactggtagccaatcgacttctaccgcttcatcaggcacctcagctagttcacgcaatactgccggttcacgtaatactgccaatccaagtaatgctgctagttcaaacaatactgctagttcaagcaatgctgctagttcaagcaatggtgccagttcaactgcaagtactgaatcgaataacgctggggaaggtgaagatgataattaagaaaataaagaaacatgacaaagataaaaataaaaataaacgttgaaaaaaaaaaaaaaaaaaaaa"
">dd_smedV4_12111_0_1" "aatttatatattaaattgaattaaacgtttaatttttatcaattttattaagttatcaaatataagtattttataaacacgagaaaatatgatttttattttcaaggatattacatttaaatttttgttggttttattgtcatcgctctattgtttttcgtcgacaatttggatcaatgatccgtctgacgaatcagaaatctgtccaaatgggtgtcatgtatgttgtctagttagttcgtttgtactctatcagtcgtacg"

[sarvidsson]

Could you post a few lines from the FASTA file you produced?

[dena.dinesh]

Could you post a few lines from the FASTA file you produced?

Hi,
I have added a few lines of my fasta file. Please take a look

Best
dena

[sarvidsson]

Posting the final file (after all replacements, cleanup etc.) would be easier to debug (e.g. as an attachment).

Make sure that the first line ("fasta_seq") is removed and that there is a linebreak between the ID and the sequence (difficult to tell whether this is the case). Additionally, some tools expect fixed-length sequence lines - you can use the "fold" command line utility to fix that.

[dpryan]

You don't want to use write.table(). Well, you can, but then you'd need sep="\n" and quote=F. A better method would be the write.fasta() command.

[dena.dinesh]

Hi Ryan,

Thanks for your comment. I tried the "write.fasta" for the file but it prints out only the first sequence with all character in a single line. Its is not printing out the other sequences. I think the file must be in different format. I have attached the file for your reference. Kindly guide me.

[dena.dinesh]

Could you post a few lines from the FASTA file you produced?

I have attached the file which was generated by above R command for your reference.

[sarvidsson]

That file should be OK (it is proper FASTA). As I previously said, some tools like to have folded sequence lines (just run the Unix command fold on it).

I ran your file on the frameDP web resource from INRA (https://iant.toulouse.inra.fr/FrameDP/), and that worked fine.

[dpryan]

The problem is that you mucked up the output of read.fasta. Your code should be something like:

library(seqinr)
ids=as.character(read.delim("path/to/ids/file.txt"))
dd=read.fasta("path/to/transcriptome.fasta",seqtype="DNA",as.string=T)
dd = dd[names(dd) %in% ids)
write.fasta(dd, names(dd), file=paste(dir,"/",name,sep=""))

There's no need to muck around with prepending ">" to the names.

[dena.dinesh]

Thanks Ryan. It worked but when I gave nbchar=70, it doesnt seems to work. rather it prints the entire sequence in a single line. Thanks once again for your help

[dena.dinesh]

That file should be OK (it is proper FASTA). As I previously said, some tools like to have folded sequence lines (just run the Unix command fold on it).

I ran your file on the frameDP web resource from INRA (https://iant.toulouse.inra.fr/FrameDP/), and that worked fine.

Thank you very much. it worked