2.0 Beta 9 having a problem about ustacks

[yang liu]

Hi Julian and Catchen ,

I was running ustacks (Beta 9) ,but it had aborted and produced a erro information [ustacks: src/ustacks.cc:983: int call_consensus(std::map<int, MergedStack*>&, std::map<int, Stack*>&, std::map<int, Rem*>&, bool): Assertion `r->seq->size() == length' failed.   Aborted (core dumped)] 

I don't know why for it

please give me help ,thanks very much

 

Yangyang

[Julian Catchen]

Hi Yangyang,

Can you share the FASTQ file that is causing the assert failure, through
a service like Dropbox?

julian

> <https://lh3.googleusercontent.com/-9dkOHboDAZ4/WqYXOOb-_ZI/AAAAAAAAAAQ/5Ryq6HAKmngozO7kSNuca_pKKMPPepKMACLcBGAs/s1600/7%2525NJP%2560%257BQK4GX8%255D2VZ%2540OK%255BQI.png>
>
>
>
>
>
> <https://lh3.googleusercontent.com/-3zojoJMAe5Y/WqYXf8wS3gI/AAAAAAAAAAU/qdn_NW7l3V4vBGKeYYzi5ssZpYc_EwmkgCLcBGAs/s1600/S_2%25603J%257BZ%25297DXMJWIL3_%2529DC8.png>
>
>

[yang liu]

Hi julian ,

It's nice to see your reply. I have baidu Dropbox , and the FASTA file of one sample was uploaded on it , link: https://pan.baidu.com/s/1XkPYyU4h_ifcjvYHu3DluQ  password：q8ce

Please use it . Good wishes.

Yangyang

[Nicolas Rochette]

Hi Yang,

It seems like I can't access the file (unless I install the Baidu app)—the error is:

BJW1.fa.gz
你下载的文件过大，请使用百度网盘客户端。
安装最新版网盘客户端
已安装新版客户端，仍无法下载，查看原因

安装完成后，重启浏览器即可高速下载
知道了

Best,

Nicolas

--
Stacks website: http://catchenlab.life.illinois.edu/stacks/
---
You received this message because you are subscribed to the Google Groups "Stacks" group.
To unsubscribe from this group and stop receiving emails from it, send an email to stacks-users...@googlegroups.com.
Visit this group at https://groups.google.com/group/stacks-users.
For more options, visit https://groups.google.com/d/optout.

[yang liu]

Hi Nicolas,

I have split the FASTA file to four parts , and  was uploaded on Google Dropbox,  https://drive.google.com/file/d/1Jk9lR825M4JZJFGNli2QxJNuPAwknoVE/view?usp=sharing

                                                                                                                                  https://drive.google.com/file/d/1qHj2_cmKWuCqCqK7sI_8vgOGuXfBEWyL/view?usp=sharing                                                         

                                                                                                                                  https://drive.google.com/file/d/1zLMhlIpFLWvxAb8c1lF-sL5JrKq58QZ6/view?usp=sharing

                                                                                                                                  https://drive.google.com/file/d/1Abtd35aMpLhwWF7D31pbM2rhpjjdPX5Z/view?usp=sharing  

You can download and merge them [cat BJW1.fa.gza* > BJW1.fa.gz] , if you like. So you should test it.

Tanks very much again.

Yangyang             

[Nicolas Rochette]

Hi Yang,

Thank you for the files. With v2.0Beta9, what I see is:

Error: different sequence lengths detected, this will interfere with Stacks algorithms, trim reads to uniform length (override this check with --force_diff_len).

So that's there problem here—50% of you reads are 146bp and 50% 151bp (did you concatenate the .1 and .2 files?!).

Also, are you sure you are using v2.0Beta9 (this error was a warning before)? Or did you specify --force_diff_len (this is not recommended)?

Best,

Nicolas

[Will Shim]

Hi Mr. Rochette,

This is Will, and I'm having the same problem as Yang. I'm just trying to do a quick preliminary run of stacks on some ddRAD sequences for several populations of my organism and see some preliminary Fst values (as a way to learn stacks). I was trying to override the different sequence lengths error (within an individual) with '--force_diff_len' to see what happens, and I put that line in the denovo_map.pl as follows:

% denovo_map.pl -T 8 -m 3 -M 4 -n 4 -o $WORK/chapter1/stacks.denovo/ --samples $WORK/chapter1/all_concat --popmap $WORK/chapter1/info/allpop.txt --force_diff_len

But my cluster spit out: Unknown command line option: '--force_diff_len'

I already concatenated all my .1 and .2 files.

Do you know how I can incorporate the  '--force_diff_len' into the denovo_map.pl pipeline?

Thanks!

Will   

[徐朋]

hi, to keep you fastq reads1 and reads2  the same length: 

##eg tranked to 120bp:

ls *fastq  | parallel -j 30 fastx_trimmer  -i {1} -o {1}.120.ff -m 120

##then discard length less than 120bp reads:

for i in `ls *ff`; do seqtk seq -L 120 $i > $i\.new; done;

##rename the new file.

It worked!

ps: to add "--force_diff_len", please edit the denovo_map.pl file:

sudo vim /usr/local/bin/denovo_map.pl

# modify :

$cmd = $exe_path . "ustacks  -t $sample

to 

$cmd = $exe_path . "ustacks --force-diff-len  -t $sample ....

(not recommented!)