Memory error with rnaSPAdes? Try QA/QC and Sub-sample sequences

[vishal_jha]

I am facing issue with rnaspades, it is

not working for some of the data

while format of all the data that I am using is corect

[jennaj]

Hello @vishal_jha

Sorry to hear you are having problems!

These are two topics that have some more general tips for the type of analysis you are doing. All assembly is very sensitive to read quality, so don’t skip QA – even if you decide not to trim the reads it is good to be aware of overall data characteristics for parameter tuning and such.

QA/QC

This topic includes a QA workflow for paired end data but could be adjusted for single end → Quality Control Start Here! multQC issue and guidance?

@jennaj

Single dataset QA example

• Hands-on: Quality Control / Quality Control / Sequence analysis

Multiple paired end dataset QA example

• Hands-on: Reference-based RNA-Seq data analysis / Reference-based RNA-Seq data analysis / Transcriptomics (see the initial steps)
• More examples → Galaxy Training!

A demonstration is in the Quality Control Q20-L20 public workflow.
How to find it →

@jennaj

• Search here GalaxyProject.org → GTN Pan-Galactic Workflow Search
• With the term Quality Control Q20-L20

Memory errors

How to accommodate memory resource errors when assembling reads → Resolving memory failures with SPAdes

@jennaj

Error message

This job was terminated because it used more memory than it was allocated.

If an assembly is running out of processing memory during execution, there are a few things you can do.

Use Case
The current work is running out of processing memory during job execution. This is different from the data storage available in your account.

What to try

• Run the reads through quality control to improve the content the tool is attempting to assemble.
• Set more input parameters to improve how the tool is attempting to assemble.
• Reduce the number of reads submitted with a tool like Sub-sample sequences. Some protocols involve pre-clustering with other strategies, too.
• Try at a different public server. Each hosts distinct resources.

---

---

Let’s start there! If you decide to send in a bug report instead, please put a link to this topic into the comments section so I can link the two up. I’m recommending that everyone try a rerun first – so if you already did that, make sure to make a comment about that so we’ll know to review closer.

Thanks!