Skip to content

5. Troubleshooting

mayofaulkner edited this page Oct 7, 2021 · 3 revisions

Both the iblapps and ibllib code bases are under active development. If you run into a problem, you might find that the issue has already been fixed! As a first step for debugging please make sure you are up to date with the latest changes on the develop branch of these repos. If you have installed the ephys alignment GUI as per the installation instructions, you can sync the latest changes to your local computer from a terminal in the following way,

cd int-brain-lab/iblapps
git pull
cd ../ibllib-repo
git pull

We have also compiled a list of issues reported in the past below. If neither of these solve the problems, please post an issue here and we will get back to you!

IBL user problems


No alf data for probe

If you get the following error message could not donwload alf data for this probe - gui will not work, the most likely cause is that the spikesorted data for that session is not on FlatIron


No histology image for subject

The naming convention used for subjects in the histology folder on FlatIron (http://ibl.flatironinstitute.org/histology/) is not always consistent with the subject names stored in Alyx. While the code attempts to account for differences in capitalisation or underscores, sometimes the naming in the histology is just too quirky. In these cases, it will be necessary to manually download the histology image required for the GUI.

The image file required is located in the downsampledStacks_25/sample2ARA/ folder of the subject on FlatIron, and has the extension .....RD.tif file. This file should be downloaded and placed in a folder called histology in the associated subject folder on your local computer

e.g. Download STD_dsIBL_WITTEN_13_200109_121739_25_25_RD.tif from http://ibl.flatironinstitute.org/histology/wittenlab/IBL_Witten_13/downsampledStacks_25/sample2ARA/

and place this file in

C:\Users\Mayo\Downloads\FlatIron\wittenlab\Subjects\ibl_witten_13\histology


External user problems


Histology and xyz_picks flipped in AP axis

Depending on the method used for registration of histology images to Allen reference atlas, after tracing in lasagna the AP coordinates are sometimes flipped. This has been reported after using brainreg software for registration. In this case the following code snippet can be used to get the correct xyz_picks.json from the xxx_pts.csv file

from pathlib import Path
from ibllib.atlas import atlas
import numpy as np
import json
brain_atlas = atlas.AllenAtlas(25)
file_track = Path(r'C:\Users\Mayo\Downloads\P01_25micron_1116364RESLICED.csv')
ixiyiz = np.loadtxt(file_track, delimiter=',')[:, [1, 0, 2]]
ixiyiz = ixiyiz[np.argsort(ixiyiz[:, 2]), :]
xyz = brain_atlas.bc.i2xyz(ixiyiz)
xyz = xyz * 1e6
xyz_picks = {'xyz_picks': xyz.tolist()}
output_path = Path(r'C:\Users\Mayo\Downloads')
with open(Path(output_path, 'xyz_picks.json'), "w") as f:
    json.dump(xyz_picks, f, indent=2)

General user problems


GUI does not display on MAC with Big Sur

On mac computers running Big Sur there have been reports that the window showing the gui does not popup. To get around this the following should be run in the terminal before launching the gui (taken from https://stackoverflow.com/questions/64833558/apps-not-popping-up-on-macos-big-sur-11-0-1#_=)

export QT_MAC_WANTS_LAYER=1