Hello!
I'm using Rockhopper for the analysis of a bacterial RNA-seq experiment. I'm not a bioinformatician, and this software constitutes, in principle, a user-friendly tool for those working with bacterial transcriptomes. The genome of the bacterial species I'm working with is not in the Rockhopper database. I have tried several things:
1) When I try to conduct the analysis by specifying the location of the folder containing the genome sequence of my bug in FASTA format and the protein file in ptt format, the software doesn't run (the analysis status bar goes up to 100% immediately and no result file is generated).
2) When I run a de novo assembly of my RNA-seq data with Rockhopper, perfectly aligned reads are less than 50%, and I obtain a file with the sequences, but no annotation. So all in all, I'm not able to determine which genes are more differentially expressed. Also the p and q values provided by the software are quite confusing.
3) If I run an analysis using the genome of a closely related species as a reference, the results are not optimal at all, as evidenced by the fact that no differential gene expression is reported for some genes that must be obviously upregulated under my experimental conditions.
The RNA quality was great (RIN values 9.8-10) and the RNA-seq data quality look also very good according to the sequence quality score plots.
Any help and/or suggestions would be highly appreciated.
Best regards,
Jonathan.
Hi Artur,
Thank you for your comment. However, the software itself requests the gene annotation file in ptt format (= protein). If the genome sequence file and/or the gene annotation files are missing or in a non-supported format, the software will tell you in a pop-up window. So all in all, yes, ptt files are not only supported but also are the only gene annotation files accepted by Rockhopper.
Hi, while working with bacterial RNA seq data I faced several problem like you are facing right now. So far edgePRO (https://www.ncbi.nlm.nih.gov/pmc/articles/PMC3603529/) gave me best result as it deals with overlapping genes. Rockhopper turns to be good predictor of operon for me. You can run Edge pro which is very easy to run in linux. Then take the read count and use Deseq2 for your differential expression analysis.
How can I get the annotation file in ptt format (= protein) to my bacteria genome?
HI, Jean I just use this perl script https://github.com/sgivan/gb2ptt which worked for me. You need to install bioperl beforehand. If you are having problem you can use xl to create the .ptt file and .rnt file. For ptt file you need first colum the coordinates like 1..100 (your gene coordinate), to create this you can use XL =CONCATENATE() option. Length should be your protein length and cog colum you can keep - if it is not available. In .rnt you should put all your ribosomal RNA and tRNA coordinate but in this case use gene length in length column. Hope you will not have any problem to generate your ptt and rnt file.
Hi A K M Firoj Mahmud,
Thank you very much for your help and comments.
Jean Luiz
We are happy to say that our lab has developped a complete package for RNA-seq data analysis for bacteria (one command for everything)! the details can be found in the link https://prokseqv20.readthedocs.io/en/latest/
Hi A K M Firoj Mahmud,
Thank you very much for your up date, I believe that will be very useful for us.
Jean Luiz
- Badges
- Science topic
- Similar topics
- Phytochemistry
- Extracts