Hello, I was recently bitten by an unexpected behavior in the sraConvert function from the SRAdb package. I wanted to fetch the other SRA IDs associated with the SRX numbers of 32 samples, and I used the sraConvert function to do so. However, I did not realized that sraConvert returns the results in arbitrary order rather than sorting them in the same order as the input, so I just used cbind to add these IDs to my existing sample table. This effectively shuffled my samples, and I did not notice until far downstream when I started making PCA plots and the clustering made no sense. I leave it up to the developers of the SRAdb package to decide whether or not this is a bug, but I think it should at least be documented that the sort order of the output of sraConvert is arbitrary and will not necessarily match the input. -Ryan
[Bioc-devel] SRAdb::sraConvert returns results in arbitrary order
2 messages · Ryan, Jack Zhu
3 days later
Hi Ryan, Thanks for your message and I will look into this. Jack ---- Yuelin Jack Zhu Biologist (Bioinformatics) Genetics Branch/CCR/NCI/NIH Tel: (301)496-4527 FAX: (301) 402-3241 E-mail: zhujack at mail.nih.gov On Thu, Nov 17, 2016 at 5:20 PM, Ryan C. Thompson <rct at thompsonclan.org> wrote:
Hello, I was recently bitten by an unexpected behavior in the sraConvert function from the SRAdb package. I wanted to fetch the other SRA IDs associated with the SRX numbers of 32 samples, and I used the sraConvert function to do so. However, I did not realized that sraConvert returns the results in arbitrary order rather than sorting them in the same order as the input, so I just used cbind to add these IDs to my existing sample table. This effectively shuffled my samples, and I did not notice until far downstream when I started making PCA plots and the clustering made no sense. I leave it up to the developers of the SRAdb package to decide whether or not this is a bug, but I think it should at least be documented that the sort order of the output of sraConvert is arbitrary and will not necessarily match the input. -Ryan
_______________________________________________ Bioc-devel at r-project.org mailing list https://stat.ethz.ch/mailman/listinfo/bioc-devel