Comments for installing the Workbench
Comments for the Workbench tools
I have a question about SiLoCo. I run it on a powerful machine which has more than 200GB of memory with the following command:
java -jar Workbench.jar -tool siloco -srna_file_list noAdptr.fasta -genome genome.fa -params siloco_params.cfg
Jun 15, 2015 3:30:54 PM uk.ac.uea.cmp.srnaworkbench.utils.WorkbenchLogger log
SEVERE: WORKBENCH: SILOCO: Message: GC overhead limit exceeded;
Stack Trace: java.lang.String.substring(String.java:1913)
Jun 15, 2015 3:31:04 PM uk.ac.uea.cmp.srnaworkbench.utils.WorkbenchLogger log
SEVERE: WORKBENCH: SILOCO: Out of Memory Error occured.
We advise increasing the amount of memory available to the JVM using the -Xmx argument or running smaller datasets through this machine.
The parameters are being given correctly,100GB of RAM should be enough to run the majority of datasets so I am surprised the GC limit is being reached. There is unfortunately no way to recover from this error it is just a case of restarting it. You could try and give it even more RAM if you have 200GB to spare.
“when I used 3.2 version on Ubuntu 12.04 everything went OK. I asked to install jdk1.8 (by Oracle) on Red Hat so specifying path to it everything worked. After a few months I had problems with 3.2 Filter (can’t find filter.seqprop.out) Switching to 3.1 worked, but miRCat on both 3.1 and 3.2 has the memory issue or it tells “Connection refused” after analysis.
miRCat error message Index: 58, Size 58 or Index 33, Size 33 etc. All this in a version 3.1 which worked reasonably good. miRCat in ver 3.2 keeps getting me this kind of error messages”
The connection refused error sometimes happens when the log files get locked and disables parts of the software. Please go into the user/logs folder and delete all of the files you find, then try again with the version that worked (3.2)
The index errors may be caused due to zombie processes hanging around after the software has crashed, you could attempt to killall of the running java processes before starting again. This could also cause the error you see with version 3.2. In addition, make sure the software has write/read privileges for the listed location.
“I have aligned my sequences with STAR to the reference genome and then extracted the sequences that aligned to miRNAs on miRBase. With the rest of the sequences I want execute miRCat to try to find new miRNAs. Is there a way to use or transform the bam file from STAR to be able to use miRCat? Or it’s mandatory to realize all the process in miRCat?”
Currently we do not have a feature for converting STAR alignment files.A straight forward option would be to perform all analysis, including the genome alignment step, in miRCat (perhaps try the newer version, miRCat2) https://www.ncbi.nlm.nih.gov/pubmed/28407097.
“I have a question regarding isomiRs. They are perfectly analysed by miRProf, but I have a lot of sequences (hairpins and their corresponding mature miRNAs) resulting from miRCat analysis. What approach would you recommend to group miRs in isomiR categories? Do you know any software which uses similar algorithm for isomiRs discovery to the one used in miRProf???”
The predicted miRNA hairpins could be used as input for miRProf i.e aligning the sequencing reads in your sample(s) to the predicted hairpins and grouping those that match to the same pre-cursor structure might be a starting point.
“I am trying to use sRNA workbench (Ver. 4.4, 4.3.1) for degradome analysis using PAREsnip but I am encountered with the following error (in screenshot). Could you please give suggest how to overcome this issue.?”