search

hyunhwan-jeong / SalmonTE

SalmonTE is an ultra-Fast and Scalable Quantification Pipeline of Transpose Element (TE) Abundances
GNU General Public License v3.0
81 stars 23 forks source link

SnakeFile Single Issue #26

Closed juanb001 closed 5 years ago

juanb001 commented 5 years ago

Hi,

I'm having an issue with single-end data on the latest version of SalmonTE and latest version of snakemake.

screen shot 2019-02-06 at 1 02 41 am
hyunhwan-jeong commented 5 years ago

@juanb001

Can you show me the list of the file names in the trimmed directory?

Thank you,

Hyun-Hwan Jeong

juanb001 commented 5 years ago

Here it is!

screen shot 2019-02-06 at 2 55 23 pm
hyunhwan-jeong commented 5 years ago

@juanb001 Thanks for reporting this. This error was because a rule had been omitted in the Snakefile. I have fixed and committed it - 600fe2e300e11dad90d5051a65a6813d6afea748. Can you pull the github repository run this again?

Thank you,

Hyun-Hwan Jeong

juanb001 commented 5 years ago

I re-downloaded and re-run, but now I'm getting this error:

screen shot 2019-02-06 at 5 06 04 pm
juanb001 commented 5 years ago

When I change directory to where my FASTQ files are (the /trimmed folder) and run SalmonTE from there, I believe it is working. However, the above issue happened when the current directory was the Desktop and I linked to the /Desktop/Senescence_Fibro_trimmed folder when I ran the program.

hyunhwan-jeong commented 5 years ago

Can you try it in the same directory with a different output directory (change sen_quant to another)?

Thank you,

Hyun-Hwan Jeong

juanb001 commented 5 years ago

By same directory, do you mean run it again from the Desktop (like in the error box) but change sen_quant to sen_quantifications (for example)?

hyunhwan-jeong commented 5 years ago

Exactly!

juanb001 commented 5 years ago

The last run (from the folder with the FASTQ files ran successfully).

However, now that I change directory to Desktop I get "Failed to read /Users/juanb/Desktop/Senescence_Fibro/trimmed/". Running from the FASTQ folder gives the same error.

juanb001 commented 5 years ago

It's working now. The problem I think was, I had the folder from the successful run in the FASTQ folder. SalmonTE had issues with that when it was reading the FASTQ files, I think. Removing the folder from the successful run made SalmonTE run again successfully from the Desktop with the new output folder. However, using the original "sen_quant" output folder keeps giving the last error shown in the picture.

hyunhwan-jeong commented 5 years ago

@juanb001 glad to hear that, but I want to figure out a couple of things.

Thank you,

Hyun-Hwan Jeong

hyunhwan-jeong commented 5 years ago

@juanb001

However, now that I change directory to Desktop I get "Failed to read /Users/juanb/Desktop/Senescence_Fibro/trimmed/". Running from the FASTQ folder gives the same error.

I am not sure you've seen this error before. If you haven't then can you make a screenshot or do a copy-and-paste the text message here?

juanb001 commented 5 years ago

The second error is:

screen shot 2019-02-06 at 6 22 59 pm

Which seems to be caused by the extra files that aren't FASTQ in the same directory as FASTQ files. If I delete sen_quant_complete, SalmonTE runs successfully, otherwise it does not:

screen shot 2019-02-06 at 6 26 23 pm

By successful run, I mean I was able to complete the quantification step.

I don't see .snakemake. I can't move sen_quant since it doesn't seem to exist. There's no folder on my Desktop with that name; terminal doesn't show it either with the ls command. If I change sen_quant -> sen_quantif the quantification goes to completion.

Finally, when I do the test function, to compare sample A to sample B I get this error. Is it because I only have one replicate for each? Are single replicates no longer supported?

screen shot 2019-02-06 at 6 35 20 pm
hyunhwan-jeong commented 5 years ago

SalmonTE uses DESeq2, and the latest version of DESeq2 no longer supports single replicates. One possible solution is downgrading the version of DESeq2, and you can do it with the following command line in R:

 install.packages("https://bioconductor.org/packages/3.6/bioc/src/contrib/Archive/DESeq2/DESeq2_1.18.0.tar.gz", repos=NULL)
hyunhwan-jeong commented 5 years ago

@juanb001

I don't see .snakemake. I can't move sen_quant since it doesn't seem to exist. There's no folder on my Desktop with that name; terminal doesn't show it either with the ls command. If I change sen_quant -> sen_quantif the quantification goes to completion.

You even did not see the folder when you use ls -alh?

hyunhwan-jeong commented 5 years ago

@juanb001 Does the DESeq2 solution help you?

juanb001 commented 5 years ago

Thanks for the help! I tried to run the code you provided but it sadly gives an error, so I can't run single replicates for now., but this is a limitation with R.

screen shot 2019-02-07 at 12 27 30 pm
hyunhwan-jeong commented 5 years ago

@juanb001 Have you installed homebrew in your machine? If then, brew install gfortran and run the code will help you.

juanb001 commented 5 years ago

I believe it's already installed, but I still got the error (on a Mac at work). I repeated the test on my computer running linux and did the single replicate comparison without any issues. Based on others' comments it seems I need to change the version of R on the mac. I'll figure it out. In any case, thanks for the help!