create the missing GFF

[fahmida]

Hi,

This could be a trivial question. I have a de novo assembled genome and predcited proteins in fasta format. However, the accompanying GFF files are missing. What's the best way of creating the GFF file? Is it: blast proteins vs genome and then something like blast2gff using the blast result?

Thanks.

[GenoMax]

You have not said anything about the size of the genome you are working with but you could perhaps use MAKER (http://gmod.org/wiki/MAKER) which will make GFF3 files.

Otherwise you can build the file yourself by parsing the information you have to fit the GFF3 format: http://www.sequenceontology.org/gff3.shtml

[fahmida]

You have not said anything about the size of the genome you are working with but you could perhaps use MAKER (http://gmod.org/wiki/MAKER) which will make GFF3 files.

Otherwise you can build the file yourself by parsing the information you have to fit the GFF3 format: http://www.sequenceontology.org/gff3.shtml

Thanks for your reply GenoMax. Assembly size is ~400mb with ~250000 scaffolds(estimated Genome size ~480mb). Maker was was used to find the genes. Output of maker was the predicted transcripts, prtoteins and their corresponding gff file. However, unfortunately we only have the predicted proteins now nad lost other files. Hence instead of running the time consuming Maker pipeline again, I was wondering is their a way of mapping/aligning the proteins aganist the genome just to create the gff file?

[GenoMax]

Thanks for your reply GenoMax. Assembly size is ~400mb with ~250000 scaffolds(estimated Genome size ~480mb). Maker was was used to find the genes. Output of maker was the predicted transcripts, prtoteins and their corresponding gff file. However, unfortunately we only have the predicted proteins now nad lost other files. Hence instead of running the time consuming Maker pipeline again, I was wondering is their a way of mapping/aligning the proteins aganist the genome just to create the gff file?

Out of curiosity how long was the original Maker run? Day(s)/week? It is a pity that you lost the predicted transcripts file otherwise this would have been easier.

It may be more work than necessary to re-map the proteins back to the scaffolds and then parse the results to make the GFF file.

[fahmida]

Out of curiosity how long was the original Maker run? Day(s)/week? It is a pity that you lost the predicted transcripts file otherwise this would have been easier.

It may be more work than necessary to re-map the proteins back to the scaffolds and then parse the results to make the GFF file.

Maker run took more than 2 weeks :-(