Suggestion: Continue download despite MD5 failure or missing accession in list mode

[zhanxiangzong]

First of all, thank you very much for providing the iseq tool — it has been very helpful for our work!

I encountered a problem during use. When downloading a batch of datasets (e.g., all CRR samples under a CRA accession), if any single sample fails MD5 check 3 times, the entire task stops instead of continuing to the next sample. Similarly, when using the list mode to download a batch of accessions, if any one accession is invalid, inaccessible, or cannot be found, the program terminates instead of skipping it.
Here is an example using iseq -i list(I added a nonexistent accession in the middle of the list) -m:

Note: You choose to skip downloading SRA files (-m used), only retrieve the metadata for each accession, see SRR28795981.metadata.tsv ==================================================SRR28795981 download finished================================================== Note: No metadata information found for SRR111111111111111 in the ENA database, try to download from the SRA database Error: No metadata information found for SRR111111111111111 in the all databases How to solve? 1. Check your accession format: SRR111111111111111; 2. SRR111111111111111 is not available in the SRA database; 3. Network is not available on your server

I’m not sure whether this is due to my local setup or a default behavior of the tool. Would it be possible to optimize this behavior or add a configurable parameter to allow skipping problematic samples and continuing with the rest?

Thank you again for the great tool!

[BioOmics]

Thank you for using iSeq. Your issue has been resolved. Please download the latest iSeq script from the following link:

🔗 https://github.com/BioOmics/iSeq/blob/main/bin/iseq

Then replace the existing version in your conda environment. You can do it like this:

$ which iseq
~/miniconda3/envs/iseq/bin/iseq
$ chmod 777 iseq
$ mv iseq ~/miniconda3/envs/iseq/bin/iseq

Since this update may have potential issues, please provide feedback after using it😀.

Once I receive your confirmation that it's stable, I will proceed to update it on Bioconda.

---

Additionally, if you prefer not to update the script, you can use the following command as an alternative:

cat list | while read i; do iseq -i ${i} -m; done

This allows you to process each item in the list without modifying the current iSeq installation.

[zhanxiangzong]

Thank you! I followed your example and replaced the script from https://github.com/BioOmics/iSeq/blob/main/bin/iseq. The issue with downloading metadata using a list has now been resolved — failed tasks no longer affect the progress of subsequent ones. However, if an md5 check failed after trying 3 times error occurs, it still interrupts the download of the entire dataset.

[BioOmics]

Hello,

The issue has been resolved after the recent update.

Please download the latest version of the iSeq script from the link below:

🔗 https://github.com/BioOmics/iSeq/blob/main/bin/iseq

Then replace the existing version in your Conda environment. You can do this as follows:

$ which iseq
~/miniconda3/envs/iseq/bin/iseq
$ chmod 777 iseq
$ mv iseq ~/miniconda3/envs/iseq/bin/iseq

Please feel free to share your feedback after trying it out 😀.

[zhanxiangzong]

Thank you, this update has resolved my issue.

[BioOmics]

Thanks. I will close this issue.

Any issues, any time you can re-open it.