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
KeyError: 'alignment_id' #151
Comments
I have verified that this is fixed on develop for both jackhmmer and hmmsearch options. Jackhmmer: I tested this using the config file @sacdallago sent this morning, which was yielding the error using the jackhmmer option. Hmmsearch: tested earlier this week with Bobby Not sure if we should close this issue as it is still something that is broken on the master branch and stable environment on o2. |
No; if a fix is merged tested in dev, we can close :) And thanks @aggreen |
This issue is fixed in the master branch of my development environment (or at least, it does not occur in my development environment). But, the issue is still occurring on evcouplings_develop on o2. I have checked the following:
return no differences between the two files (my environment package vs development environment package).
So, I'm really at a loss as to what could be causing this problem, and as of now I can't say that the development branch is really working. I'm attaching a culprit config file if anyone would like to try to reproduce this issue. Are we sure that the evcouplings_develop is really using the evcouplings found in the site packages folder? That's the only other possibility I can think of. |
@aggreen I reproduced the same problem on o2 with the evcouplings_develop environment. I am a bit confused though... I don't see any recent commits of yours in here: https://github.com/debbiemarkslab/EVcouplings/commits/develop/evcouplings/compare/sifts.py Could you point me to the commit where you fixed the issue? |
I believe this was fixed by the January 2nd commit, c2a166f Can you reproduce the lack of a problem with my environment, agg_evcomplex? This may be a new error which I can't reproduce for whatever reason, but it seems like the same issue as was fixed before. |
Seems to be some sort of pandas version issue... the by_alignment() call worked successfully on my local machine with pandas 0.19 until I upgraded to 0.22, now I get the same problem as on evcouplings_develop, which is on 0.22. Your agg_complex environment is on 0.20.3... I'll have a look now if I can trace down why this breaks all of a sudden with the latest version of pandas. |
This is a groupby agggregation gone wrong - dataframe is empty cause no structures found, index columns are lost during aggregation because of empty dataframe, and then missing during the subsequent merge. Allegedly, this issue pandas-dev/pandas#8093 has been fixed in pandas 0.22 but I find it odd that this is the very release that first caused the problem for us. Anyways, fixed now and merging into develop in a second. |
Should be working now on any pandas version (I haven't updated the evcouplings_develop environment yet). Closing for now, please reopen if this resurfaces ever again. |
Thanks a lot Thomas. I will keep an eye out, hopefully it will not
resurface in the future.
…On Fri, Mar 30, 2018, 11:08 AM thomashopf ***@***.***> wrote:
Closed #151 <#151>.
—
You are receiving this because you were mentioned.
Reply to this email directly, view it on GitHub
<#151 (comment)>,
or mute the thread
<https://github.com/notifications/unsubscribe-auth/AHimXhP50b4uEkcf9hOnbDuS-Q2wvemvks5tjkpzgaJpZM4S5vvk>
.
|
This problem happens on stable in compare and makes the execution of later stages crash.
AFAIK it happens when no PDB structure to compare with is found, a complete stack follows.
In the config for this file, the pdb alignment method is set so:
@aggreen should have fixed this on develop but due to the circular dependency, it's hard to test, and I would keep it open until confirmed fixed. Also, I remember @aggreen saying yesterday that it's odd that it also happened using
jackhmmer
, so just for the reference.The text was updated successfully, but these errors were encountered: