Skip to content
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.

Sign up for GitHub

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

Jump to bottom

Use of uninitialized value $esearch_result in pattern match (m//) #1

Closed
ckeuschn opened this issue Mar 14, 2018 · 23 comments
Closed

Use of uninitialized value $esearch_result in pattern match (m//) #1

ckeuschn opened this issue Mar 14, 2018 · 23 comments

Comments

@ckeuschn
Copy link

Hi,

I successfully installed SACCHARIS and additional software. I tried to follow the example from your publication to test the pipeline and got an error message after the cazy_extract.pl script.

user$ cazy_extract.pl -f GH42 -g characterized
Use of uninitialized value $esearch_result in pattern match (m//) at /usr/local/Cellar/SACCHARIS/cazy_extract.pl line 462.
Use of uninitialized value $count in concatenation (.) or string at /usr/local/Cellar/SACCHARIS/cazy_extract.pl line 465.
Use of uninitialized value $esearch_result in pattern match (m//) at /usr/local/Cellar/SACCHARIS/cazy_extract.pl line 469.
Use of uninitialized value $esearch_result in pattern match (m//) at /usr/local/Cellar/SACCHARIS/cazy_extract.pl line 472.
Use of uninitialized value $key in concatenation (.) or string at /usr/local/Cellar/SACCHARIS/cazy_extract.pl line 476.
Use of uninitialized value $web in concatenation (.) or string at /usr/local/Cellar/SACCHARIS/cazy_extract.pl line 476.
Use of uninitialized value $efetch_result in substitution (s///) at /usr/local/Cellar/SACCHARIS/cazy_extract.pl line 482.
Use of uninitialized value $_ in split at /usr/local/Cellar/SACCHARIS/cazy_extract.pl line 165.

apparently $esearch_result is not assigned with a value (if I understand correctly), so does that mean the URL submission to the Entrez API fails to get any sequences? Maybe you could help me to solve that problem?

Thanks!
@DallasThomas
Copy link
Owner

DallasThomas commented Mar 15, 2018 via email

@ckeuschn
Copy link
Author

Thanks for your help

I was thinking that it might also be that the $id_list object has no assigned value due to some problem during pulling ids from the cazy database earlier in the code

Christoph

@DallasThomas
Copy link
Owner

DallasThomas commented Mar 15, 2018 via email

@ckeuschn
Copy link
Author

The URL which is created seems to be fine, at least the code still matches the structure of the URL of cazy.org

I attached a file with the output of the Saccharis.pl script, maybe this helps to clarify things.

Thanks again
Saccharis_error.txt

@DallasThomas
Copy link
Owner

DallasThomas commented Mar 20, 2018 via email

@DallasThomas
Copy link
Owner

DallasThomas commented Mar 26, 2018 via email

@ckeuschn
Copy link
Author

Hi Thomas,

Yes I changed line 236 in the TagParser.pm as you said and I get the same error messages even when I just enter one attribute in the -g option.

I will be traveling the next two weeks, but I will try to answer as soon as possible during that time....
Thanks
Ch

@DallasThomas
Copy link
Owner

DallasThomas commented Mar 26, 2018 via email

@DallasThomas
Copy link
Owner

DallasThomas commented Mar 26, 2018 via email

@cmorganl
Copy link

Hi Dallas and Christoph,

I hope you don't mind me piggy-backing on this issue but I'm encountering a similar error probably stemming from the same problem. While I wasn't able to reproduce Christoph's error with perl cazy_extract.pl -f GH42 -g characterized, I was able to get similar errors with:
perl cazy_extract.pl --family GH109 --group all. I'm running RedHat with perl v5.16.3 built for x86_64-linux-thread-multi and HTML::TagParser 0.20. I hope this helps.

Additionally, the command perl cazy_extract.pl --family GH3 --group all resulted in:

Use of uninitialized value $num in multiplication (*) at cazy_extract.pl line 259.

and an empty FASTA file. Any help would be appreciated!

Thanks,
Connor

@DallasThomas
Copy link
Owner

Hello Connor,

Thanks for the input. I have uploaded a new version of cazy_extract.pl that fixes the $num error. This is common and has to do with style changes that happen on the cazy website every once and a while.

As for the error with GH109 I am still tracking that one. NCBI esearch generally allows a query of 500 sequences. I have a feeling this has changed and would totally explain the errors.

Christoph - are you getting the errors with GH42 characterized as well as all when run separately?

If what I suspect is happening based on an update from NCBI and the use of their esearch tools. I will be in contact with them to determine a work around. I will keep everyone posted.

I apologize for this.

Dallas

@cmorganl
Copy link

Hi Dallas

Thanks for the quick push! It looks like it solved the problem when searching for all sequences, but when I choose a taxonomic group the $num error persists. Here's the command I used:

perl cazy_extract.pl --family GH3 --group bacteria

Thanks!
Connor

@DallasThomas
Copy link
Owner

Hello Connor,

Similar issue to before - updates have been made to cazy_extract.pl - please keep me informed if you find more of these.

I did email NCBI and am just waiting to hear back from them.

@ckeuschn
Copy link
Author

Hey Thomas,

cazy_extract.pl without arguments gives no errors. My system is actually Mac OS X El Capitan and I run perl v5.26.1. and HTML::TagParser 0.20 and LWP::Simple 6.33.
I might try to run the scripts also from a virtual Linux distribution when I am back to see if that does the trick.
Thanks
Ch

@DallasThomas
Copy link
Owner

DallasThomas commented Mar 27, 2018 via email

@DallasThomas
Copy link
Owner

Connor,

I have just uploaded a new version of cazy_extract.pl that deals more efficiently with the previous $num errors and should cover other potential cases.

Dallas

@cmorganl
Copy link

Hi Dallas,

I don't think this new version is working properly. It printed some variables then exited.

$ perl cazy_extract.pl --family GH3 --group characterized
$VAR1 = '100';
$VAR1 = 3;

Connor

@DallasThomas
Copy link
Owner

Ha, whoops I am an idiot - forgot to remove my debugging stuff. Just removed it - this version will work. In your version the problem lines are around line 255 - the two print Dumper lines and the exit. You can either download this version or just remove those lines from yours. Sorry about that.

@cmorganl
Copy link

Haha, no worries. I figured that was what happened... I'm also guilty of this.

Thanks again!

@DallasThomas
Copy link
Owner

Hey Christoph and Connor,

There was a change with NCBI and a further change coming from NCBI in the next couple of months. I have fixed the first issue in the current version so there should be no more issues as first reported by Christoph.

The second change will be implemented in a new version I should have up in a couple of days. NCBI is starting to experience increased draw on their servers, to prevent over subscription they are placing some checks in place to prevent this. The current implementation will be all users will need to submit a unique name for their program (the name is just for their tracking so can be whatever you want) and email address. Users will also have to create an account with NCBI which will give them a unique API key. Program names, email and API keys will be required for the e-utilities.

How this will effect the cazy_extract.pl script is in the new version once downloaded all users will have to first register Name and email with Cazy by sending an email to eutilities@ncbi.nlm.nih.gov - users will also have to register an account with NCBI and get their API key from the settings page. Program name registered should have no spaces. The user will then have to modify cazy_extract.pl - by first un-commenting the variable declarations for email, name and API key and then entering in single quotes those values.

This version as I said should be up by the end of the week. I will send another comment once it is available.

Thanks
Dallas

@DallasThomas
Copy link
Owner

Sorry for the delay.

The modified version with the NCBI registration has been uploaded. Please see the section of the help page for NCBI registration. It outlines what is needed from NCBI and where cazy_extract.pl will need modified by the user.

Christoph - have you had any luck yet?

@cmorganl
Copy link

Hey Dallas,

After the fresh git pull, I completed the 3 steps to set up an account to access NCBI E-utils. After removing an exit call from line 41, everything works perfectly now.

Thanks a bunch!
Connor

@DallasThomas
Copy link
Owner

Thanks for the heads up on the exit call. My bad.

@DallasThomas DallasThomas mentioned this issue Apr 5, 2019
Closed
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Assignees
No one assigned
Labels
None yet
Projects
None yet
Milestone
No milestone
Development

No branches or pull requests
3 participants
@DallasThomas @cmorganl @ckeuschn