Frequently Asked Questions
How to open the results once downloaded?
- A .csv file with the ranked list of candidates: should open with Excel, OpenOffice or any spreadsheet editor.
- A .fasta file with the protein sequences of the candidates: any text editor will open this.
- A .txt file with all the RPS-BLAST results: any text editor will do the job.
How can I interpret the results?
What do the different Length Scoring Models correspond to?
The archive I've downloaded is empty. What's the problem?
A candidate I believe to be a T/A is colored in orange in the result table. Why?
When I click on an anchor link to reach a possible partner, nothing happens...
Can I use RASTA-Bacteria with any prokaryotic sequence or only with eubacteria?
Unzip the .tgz package: any basic compression utility should do. Alternatively you can open a terminal window, set the current working directory to where you've downloaded the tarball (e.g. 'cd RASTA_dwnlds/') and type 'tar -xvzf XXX.tgz'. This will create a directory with 3 types of files:
All the Open Reading Frames (ORFs) found in a sequence, whether annotated or not, undergo a scoring process to determine their likelihood to belong to the T/A family of genes, and are ranked accordingly (in descending order). Scores above 70% are pretty sure, and the range 70-55% constitutes a twilight zone in which a candidate should be carefully verified before being discarded, given that it bares a T/A feature. Nevertheless, always keep in mind all features should be considered before selecting an ORF as a T/A.
Each length scoring model was calculated on the basis of the lengths of the toxin/antitoxin proteins reported in the broad study by Pandey et al. (2005) available for the corresponding sub-group of prokaryotes.
In most cases, this is due to the sequence submitted not being in a valid format. Please ensure you only input fasta sequences, or Genbank annotated sequences. Also, for pasted sequences as well as uploaded personal files, please note that you can only submit one sequence per job at a time. This is for data transfer reasons, and it might change in the future, however at the moment the interface is not designed to cope with multiple sequences.
The background colors in the 'Score' column of the result tables are simple indicators. They have been added to give the user a sense of the score ranges in which candidates have most chances to reflect a biologic reality. Thus green represents broadly reliable candidates; orange requires double checking (each feature should be verified); red designates low confidence candidates, most likely belonging to other families of genes.
This simply means the confidence score of the ORF pointed by the link was not high enough for it to be ranked
in the top 200 ORFs, and thus displayed in the web browser. If you have reasons to believe this "ghostly neighbour" has
been mis-scored, then the best thing to do is download the tar archive corresponding to your results and check the
data locally in your favorite table editor.
(Nb: Reaching a neighbour thanks to an anchor link requires the use of single-line tables, which explains why gaps
sometimes occur in the columns.)
RASTA-Bacteria can process both archaeal and eubacterial genomes. We just thought "RASTA-Bacteria" sounded better than "RASTA-Prokaryote"!! :-) In fact, the parsing of the data leads the tool to tolerate eukaryotic sequences as well, although no eukaryotic T/A has yet been reported (as of august 2007).
 |
Page last modified: March 08, 2020. |  |