Entering edit mode
Walter F. Baumann
▴
10
@walter-f-baumann-12439
Last seen 7.0 years ago
Hi,
I want to use sleuth on the data salmon returned. In the first step, because sleuth needs actually kallisto, I used wasabi to generate the abundance.h5 file. However, after using
so <- sleuth_prep(s2c, ~ condition, num_cores = 1)
(s2c looks like that: )
sample condition 1 sample1 D 2 sample2 D 3 sample3 D 4 sample4 C 5 sample5 C 6 sample6 C path 1 path/to/folder/sample1 2 path/to/folder/sample2 2 path/to/folder/sample3 2 path/to/folder/sample4 2 path/to/folder/sample5 2 path/to/folder/sample6
I get the following error:
reading in kallisto results dropping unused factor levels ...... normalizing est_counts 117728 targets passed the filter normalizing tpm merging in metadata summarizing bootstraps .Error in process_bootstrap(i, samp_name, kal_path, num_transcripts, est_counts_sf[[i]], : File h5 has no bootstraps.Please generate bootstraps using "kallisto quant -b".File Y:/AG_Huebner/Eric/tmp_for_jobs/seq_polyA_trim/hs_li_001_D_mR_tr_clean.fastq.gz_quant/abundance.h5 has no bootstraps.Please generate bootstraps using "kallisto quant -b".
My session Info:
Session info -------------------------------------------------------------------------------------------- setting value version R version 3.4.2 (2017-09-28) system i386, mingw32 ui RStudio (1.1.383) language (EN) collate English_United States.1252 tz Europe/Berlin date 2017-11-07
sleuth and wasabi are not Bioconductor software, and so this support site isn't the best place to post questions about use of these software.
Here is the support forum for Sleuth:
https://groups.google.com/forum/#!forum/kallisto-sleuth-users
And you could post wasabi questions here:
https://github.com/COMBINE-lab/wasabi
Hello Walter,
I am trying to adopt the salmon > wasabi > sleuth pipeline and have just run into the very same problem. I've started the topic in wasabi github site. Feel free to join there. Would be great to solve this issue and move on.
Best,
Karel
Hi, first of all thanks for moving this discussion. There is a bootstrap mode (
--numBootstraps)
in salmon and this added information that is needed by wasabi. I set it on 10. The bootstraps are needed to estimate the technical variance in your sample, which is needed by sleuth (and I guess this is what makes sleuth special compared to DESeq2 and edgeR) (according to the manual).