sdecesco / targetdb Goto Github PK

Sending for fpocket analysis only the crystal structures (or NMR/Cryo-EM) that satisfies certain criteria of quality and resolution

It looks like requests and sklearn are missing from the dependencies. I installed targetDB with pip to a clean virtual environment based on the instructions and it complained of missing these packages until I added them, just by doing:

pip install requests sklearn

Currently TargetDB can only read old-style PDB files. However, many recent depositions into the PDB are mmCIF only.

To ensure that TargetDB is able to perform pocket analyses on all structure available, support for mmCIF is needed.

I have not been able to get "List mode" to work. I load the Tkinter UI, put in a few genes (e.g. KRAS, BRAF) select "List mode" and hit "Start". I get an error message from pandas:

(snip)

File "/tmp_mnt/filer1/unix/psaffrey/.venv-targetdb/lib/python3.7/site-packages/targetDB/target_descriptors.py", line 658, in <lambda>
    lambda x: max_range(x, dx=0.1)).reset_index().rename(
  File "/tmp_mnt/filer1/unix/psaffrey/.venv-targetdb/lib/python3.7/site-packages/targetDB/target_descriptors.py", line 294, in max_range
    x.index = x.target_id
  File "/tmp_mnt/filer1/unix/psaffrey/.venv-targetdb/lib/python3.7/site-packages/pandas/core/generic.py", line 5500, in __setattr__
    return object.__setattr__(self, name, value)
  File "pandas/_libs/properties.pyx", line 70, in pandas._libs.properties.AxisProperty.__set__
  File "/tmp_mnt/filer1/unix/psaffrey/.venv-targetdb/lib/python3.7/site-packages/pandas/core/generic.py", line 766, in _set_axis
    self._mgr.set_axis(axis, labels)
  File "/tmp_mnt/filer1/unix/psaffrey/.venv-targetdb/lib/python3.7/site-packages/pandas/core/internals/managers.py", line 216, in set_axis
    self._validate_set_axis(axis, new_labels)
  File "/tmp_mnt/filer1/unix/psaffrey/.venv-targetdb/lib/python3.7/site-packages/pandas/core/internals/base.py", line 58, in _validate_set_axis
    f"Length mismatch: Expected axis has {old_len} elements, new "
ValueError: Length mismatch: Expected axis has 691 elements, new values have 649 elements

Have I done something wrong? I can get output from these genes using "Single mode".

While testing the Conda package (Im hopefully nearly there). I found syntax errors, it seems you are returning a print statement. While it won't affect the Conda package it may be be best to fix and make a new pypi release so the Conda package is up to date.

Affected lines are here:

Hello, there seems to be an issue with opentargets when trying to install the package (as is mentioned in the README as well). I was wondering if it is possible to get an older working version, or any other fix to this problem?

As of now, if a fix is not possible but it is possible to use a subset of features such as obtaining the "safety score", that does not depend on OpenTargets, that would also be amazing.

The current version of targetDB fails to run on macOS (tested on High Sierra and Mojave with anaconda 3 with python 3.7). When attempting to run targetDB, it terminates with an error similar to this one:

2019-07-18 09:21:52.503 python[9168:493584] -[NSApplication _setup:]: unrecognized selector sent to instance 0x7ff488506d00
2019-07-18 09:21:52.508 python[9168:493584] *** Terminating app due to uncaught exception 'NSInvalidArgumentException', reason: '-[NSApplication _setup:]: unrecognized selector sent to instance 0x7ff488506d00'

This is matplotlib-related and can be fixed by adding a few lines of code to targetDB/utils/targetDB_gui.py:

from tkinter.filedialog import askopenfilename, askdirectory

# start fix for macOS
from sys import platform
if platform == 'darwin':
    import matplotlib
    matplotlib.use('TkAgg')
# end fix for macOS

import matplotlib.pyplot as plt

and the same fix in targetDB/target_descriptors.py:

import scipy.stats as sc

# start fix for macOS
from sys import platform
if platform == 'darwin':
    import matplotlib
    matplotlib.use('TkAgg')
# end fix for macOS

import matplotlib.pyplot as plt

With these fixes in place, targetDB runs without problems.

Hi Stephane,

I found this tool through Benedikt Kessler and it looks cool. I was hoping to try and run it on one of our projects.

Despite not installing it yet (I will get around to that soon), I was wondering if it would be possible to update the running instructions so they are a little more detailed?

Also, do you intend on releasing this as a Conda package? Im happy to help if needed.

Adam

In my case, the main use-case for TargetDB is the following:

• Input a list of genes
• Run TargetDB in list mode
• Review the analysis as generated

It would be great if we were able to run TargetDB in a non-GUI mode, since this would allow me to deliver analysis decoupled from the Tkinter frontend. Do you have any suggestions on how we might do this?

At the moment I'm unable to install this using the supplied instructions. Installing with pip gives an error relating to opentargets, as you mention in the README:

$ pip install targetDB
(snip)
ERROR: Cannot install targetdb==1.0.0, targetdb==1.0.1, targetdb==1.0.2, targetdb==1.1.0, targetdb==1.1.1, targetdb==1.1.2, targetdb==1.1.3, targetdb==1.1.4, targetdb==1.1.6, targetdb==1.1.7, targetdb==1.2.0, targetdb==1.2.1, targetdb==1.3.0, targetdb==1.3.1 and targetdb==1.3.2 because these package versions have conflicting dependencies.

The conflict is caused by:
    targetdb 1.3.2 depends on opentargets
(snip)

You can see that pip actually downloaded and attempted to install all previous versions of targetdb without success because of the missing opentargets dependency.

I also tried installing using the conda instructions and this seemed to fail at an earlier stage:

$ conda create --name targetdb --channel bioconda targetdb
Solving environment: failed

PackagesNotFoundError: The following packages are not available from current channels:

  - targetdb
  - fpocket
  - targetdb
  - opentargets
  - addict
  - targetdb
  - opentargets
  - hyper

I have updated my targetDB installations on macOS Catalina and Ubuntu 18.04 and was able to run it in single mode, but when I run it in list mode, I get the following errors and it then hangs until I exit with ctrl-C. Any clues what might be wrong?

[NAME CONVERSION]: Converting gene names into IDs (Uniprot,Ensembl,HGNC) [NAME CONVERSION]: Conversion Done Exception in Tkinter callback Traceback (most recent call last): File "/Users/sterk/anaconda3/lib/python3.7/tkinter/__init__.py", line 1705, in __call__ return self.func(*args) File "/Users/sterk/anaconda3/lib/python3.7/site-packages/targetDB/utils/targetDB_gui.py", line 396, in launch dr.get_list_excel(self.gene_df) File "/Users/sterk/anaconda3/lib/python3.7/site-packages/targetDB/druggability_report.py", line 911, in get_list_excel data = data[col_order] File "/Users/sterk/anaconda3/lib/python3.7/site-packages/pandas/core/frame.py", line 2934, in __getitem__ raise_missing=True) File "/Users/sterk/anaconda3/lib/python3.7/site-packages/pandas/core/indexing.py", line 1354, in _convert_to_indexer return self._get_listlike_indexer(obj, axis, **kwargs)[1] File "/Users/sterk/anaconda3/lib/python3.7/site-packages/pandas/core/indexing.py", line 1161, in _get_listlike_indexer raise_missing=raise_missing) File "/Users/sterk/anaconda3/lib/python3.7/site-packages/pandas/core/indexing.py", line 1252, in _validate_read_indexer raise KeyError("{} not in index".format(not_found)) KeyError: "['Male tissues', 'Muscle tissues', 'Female tissues', 'Bone marrow & lymphoid tissues', 'Endocrine tissues', 'Adipose & soft tissue', 'Liver & gallbladder', 'Gastrointestinal tract', 'Kidney & urinary bladder', 'Proximal digestive tract'] not in index"