-
Notifications
You must be signed in to change notification settings - Fork 2
New issue
Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.
By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.
Already on GitHub? Sign in to your account
Unable to run simulated human HiFi reads #138
Comments
Hi @Oieswarya, Is that the full standard out and error? After Can you confirm you are using exactly the same environment and installation as past runs? Do you see the help page when you run Thank you for your interest in GoldRush! |
Also having a similar issue trying to run goldrush, ubuntu 20.04, 22.04, WSL2, all machines have the same error when installed with conda.
I haven't been able to build it from source yet because of missing shared libraries that I can't figure out. |
Hi @th-of, This looks like a different error/issue - would you mind opening a new GitHub issue so we can keep our discussions separate? In particular, we would want to see your command and full log (standard out and error), as well as the result of running our assembly demo. |
As I was reproducing my issue I found the problem, I was including the file extension in the reads name ("reads=reads.fastq"). However, that only changed one problem to another (see below). I will spend some more time trying to fix it before I make an issue for this one.
|
Sounds good, @th-of. I'll look out for your fresh issue. |
All is working now! Although one of the steps in the goldrush pipeline (goldpolish?) appears to be incompatible with fastq files from Dorado. The formatting of the header line seems to be the problem. The fastq file causes an error with header:
If I rename all the headers in the fastq file to a single word it runs without problems. Probably doesn't account for a tab-separated list as a fastq header. Dorado generates this by default when basecalling ONT pod5 files to fastq. |
Glad it's working for you now! Huh strange - We've tested reads from Dorado before, but perhaps not with this header format. Thanks for that info, we'll take a look at fixing that. |
Hi @lcoombe, yes strangely I have not changed anything and also goldrush-path --help gives me all the information from the help page. I am using the job script that I used to submit my previous jobs. I am using 370GB memory which should be more than enough, but do you think it is a memory issue? I also checked the headers of my fastq file and they are single words like @1 and so on. |
Thanks for confirming, @Oieswarya! Looking at your command, your target genome looks to be ~3Gbp, so yes that should be enough memory. In the
It would be super helpful to get more log messages from that command - it would be strange to just immediately fail without writing any of it's regular messages to log, if the binary itself seems OK (as indicated by you seeing the help page just fine) |
@lcoombe I wanted to update you. I have run the command separately from goldrush_immediate_files and this is the log file that generated: |
Hi @Oieswarya, So that indicates that the run went just fine (I don't see any errors), so unsure why you were getting that error before? |
@lcoombe shall I run this command now? goldrush run reads=Human_nonACTG_fq G=3120e6 track_time=1 m=10000 --debug |
That's right! Fingers crossed it'll work - if so, it could have been a transient server issue. |
I am still getting the same error: When I ran the goldrush-path command, though it was running but I did not see any files in the folder nor any soft links which it usually produces. |
Are you running that command in the same folder? It doesn't appear to be starting in the right place (ie. it is re-running the goldrush-path command) - but regardless, I can't really see any error there - could you attach the full log files to GitHub? In addition, could you re-run a fresh demo with your current set-up, just to make sure that nothing happened with your environment or server that you're usingg? |
Yes I am running both the commands from the same environment where I installed my goldrush. I will try to upload the file but unsure if I can do that as it is a 61gb file. I will run goldrush with another file (which previously successfully ran) and see if there is something wrong with the installation somehow. Thank you for your prompt responses! |
No worries! Just to clarify - I was asking about you sharing your full log files from the failed run, not your reads :) And I know you're using the same environment, but always good to do that fresh demo run as a sanity check - running a previously successful read set works too! |
@lcoombe I apologize for not getting back to this sooner. I got busy with some other work. But I tested with an input which previously ran successfully and I am having the same issue with that one too. I do not know if something happened to the installed paths but maybe I can re-install. In that case, can you guide me on how do I first delete my previously installed version? |
Hi @Oieswarya, Yes, if a sample that worked before for you isn't working now, I would suggest re-installing GoldRush. If you are using conda, the easiest way to do this would be to make a fresh conda environment, and install GoldRush there. No need to delete the previous environment, just make sure that the new one is activated when you launch your GoldRush run. Also, when you launch the job, do make sure that you are in a totally new and clean directory, just to make sure that existing intermediate files are not interfering with things. |
Thank for the prompt reply @lcoombe. I will do that and will keep you updated. |
Hello,
I have been trying to run goldrush with simulated HiFi reads of Human. The coverage of the reads is 10x. I have used goldrush for several other simulated inputs and it ran. I also checked if there is any non actg characters on my fq file and found none.
I have used this command:
goldrush run reads=Human_nonACTG_fq G=3120e6 track_time=1 m=10000 --debug
This is the .out file:
GNU Make 4.3
Built for x86_64-conda-linux-gnu
Copyright (C) 1988-2020 Free Software Foundation, Inc.
License GPLv3+: GNU GPL version 3 or later http://gnu.org/licenses/gpl.html
This is free software: you are free to change and redistribute it.
There is NO WARRANTY, to the extent permitted by law.
Reading makefiles...
Updating makefiles....
Updating goal targets....
File 'run' does not exist.
Must remake target 'run'.
mkdir -p goldrush_intermediate_files
cd goldrush_intermediate_files && ln -sf ../Human_nonACTG_fq.fq && goldrush run-in-dir reads=Human_nonACTG_fq G=3120e6 t=48 z=1000 track_time=1 k=22 w=16 tile=1000 b=10 u=5 a=1 o=0.1 x=10 h=3 s=1011011110110111101101 m=10000 M=5 r=0.9 P=15 d=5 span=2 dist=500 k_ntLink=40 w_ntLink=250 rounds=5 polisher=goldpolish polisher_mapper=minimap2 shared_mem=/dev/shm
GNU Make 4.3
Built for x86_64-conda-linux-gnu
Copyright (C) 1988-2020 Free Software Foundation, Inc.
License GPLv3+: GNU GPL version 3 or later http://gnu.org/licenses/gpl.html
This is free software: you are free to change and redistribute it.
There is NO WARRANTY, to the extent permitted by law.
Reading makefiles...
make[1]: Entering directory '/home/goldrush_intermediate_files'
Updating makefiles....
Updating goal targets....
File 'run-in-dir' does not exist.
File 'check-G' does not exist.
Must remake target 'check-G'.
Successfully remade target file 'check-G'.
File 'check-reads' does not exist.
Must remake target 'check-reads'.
Successfully remade target file 'check-reads'.
File 'clean' does not exist.
File 'goldrush_asm_golden_path.fa' does not exist.
File 'goldrush_asm_silver_path_all.fq' does not exist.
File 'goldrush_asm_silver_path_5.fq' does not exist.
Must remake target 'goldrush_asm_silver_path_5.fq'.
command time -v -o goldrush_asm_silver_path_5.fq.time goldrush-path -k 22 -w 16 -t 1000 -u 5 -a 1 -o 0.1 -p goldrush_asm_silver_path -i Human_nonACTG_fq.fq -h 3 -j 48 -x10 -P 15 -d 5 -s 1011011110110111101101 -g 3120e6 -b 10 -r 0.9 --silver_path -M 5 -m 10000 --verbose
make[1]: Leaving directory '/home/goldrush_intermediate_files'
This is the .err file:
make[1]: *** [/home/.conda/envs/goldrush_env/bin/goldrush.make:251: goldrush_asm_silver_path_5.fq] Error 127
make: *** [/home/.conda/envs/goldrush_env/bin/goldrush.make:203: run] Error 2
Can you kindly guide me as to where I am going wrong.
The text was updated successfully, but these errors were encountered: