Welcome to expipe-plugin CINPLA’s documentation!¶
Expipe is a python module for neuroscientific data analysis.
Installation¶
Windows¶
Download and install Atom Download and install Anaconda Download and install Github desktop
In CINPLA on windows, we recommend to put all expipe related softare in
c:\apps\.
General¶
We recommend installing the expipe dev package found at https://github.com/CINPLA/expipe-dev, clone with your favorit git software, on the command line write:
$ git clone https://github.com/CINPLA/expipe-dev.git
Due to dependency issues we have to make two separate environments, one for phy and one for expipe with browsers. These environments will hopefully be possible to merge in the near future.
In short, install an expipe environment with browsers, and a phy environment for spikesoting, remember that you can not copy paste everything in to the command line and press enter - copy paste and press enter on one by one line:
cd c:\apps\expipe-dev
conda env create -f expipe-environment.yml
activate expipe
python setup.py develop --extra browser
conda env create -f phy-environment.yml
activate phy
python setup.py develop --extra phy
Configuring expipe¶
Now that you have the enviroments installed you need to configure expipe. In
c:\apps\expipe-plugin-cinpla/utils you will find example configuration files.
Start by copying all files to c:/users/uiousername/.config/expipe.
Open config.yaml with Atom, edit according to your user information.
Remove processing if you do not know what it is for.
data_path: c:/users/uiousername/expipe_temp_storage
processing:
data_path: /home/user/expipe_temp_storage
username: <processing server username>
hostname: user@ipaddress
norstore:
data_path: /norstore_osl/projects/NS9048K/server
username: <norstore username>
hostname: login.norstore.uio.no
firebase:
email: your@email.com
password: yourpassword
config:
apiKey: AIzaSyAjGqZwiCKS2333m820e9UdZ7jbnkfEpjw
authDomain: expipe-26506.firebaseapp.com
databaseURL: https://expipe-26506.firebaseio.com
storageBucket: expipe-26506.appspot.com
Here we have defined two servers, one for norstore and one for a processing server e.g. your office computer; see [usage](address to usage). Note that you can add arbitrary names for servers.
Following the instructions you should be be able to:
expipe --help
Examples¶
Nomenclature for CINPLA¶
Axona
FILENAME: “axona navn” e.g. 22031701, DATE: 220317, SESSION:
01, SUBJECT: subject-id e.g. 1704, ACTION-ID is thus SUBJECT-DATE-SESSION.
Open Ephys
FOLDERNAME: e.g. 1704_2017-05-13_13-52-26_03, DATE:
2017-05-13_13-52-26, which is automatically shortened to 130517, SESSION:
03, SUBJECT: subject-id e.g. 1704, ACTION-ID er thus SUBJECT-DATE-SESSION
NWB file structure and exdir
An action is stored in a folder named after the ACTION-ID. In this folder
you will find main.exdir containing acquisition, general, analysis,
processing. The acquisition contains all the raw data from your measuring
instrument, processing contains extracted data which is and shall be processed.
Finnaly, in the analysis folder, you will find figures etc. after analysis
is perfomed.
General¶
Note that you can get all available commands with:
$ expipe --help
Furthermore if you want help on one particular command:
$ expipe transfer --help
Transfer data to and from server¶
Currently we support two means of transferring data: SCP (SSH), and copy.
To maximize transfer speed we recomend using SCP (expecially on Windows). The
SCP protocoll in expipe-plugin-cinpla is governed by the command transfer.
In the config.yaml file (see installation page) you will find:
data_path: c:/users/uiousername/expipe_temp_storage
norstore:
data_path: /norstore_osl/projects/NS9048K/server
username: <norstore username>
hostname: login.norstore.uio.no
Here norstore is the server name, the first data_path is the local path and
the data_path under norstore we refer to as the server path. To transfer
local data to the server, in it’s most simple form the transfer command can be
used as:
$ expipe transfer ACTION-ID --from-local
This is because norstore is the default server name. In order to transfer
to a different server we need to add it’s name with the --server command:
$ expipe transfer ACTION-ID --from-local --server processing
When you want to retrieve your data for manual analysis, you don’t want to
download all the acquisition data, e.g. only processing and
analysis. This can be achieved with the --exclude and --include:
$ expipe transfer ACTION-ID --to-local --include processing --include analysis
Note that many commands have a short name, e.g. --include can also be used
with -i:
$ expipe transfer ACTION-ID --to-local -i processing -i analysis
Conversly if you want to transfer the entire action only excluding the acquisition:
$ expipe transfer ACTIO-ID --to-local --exclude acquisition
Registering subject¶
Subjects are registered under the project-id subjects-registry. This is to
have a complete overview of animals in the lab i.e. active, dead, newborn
which information is stored in tags. To register a subject use register-subject:
$ expipe register-subject SUBJECT --birthday 01.02.2017 --cell_line wild-type
Registering surgery¶
When performing surgery you can register information with the register-surgery
command.:
$ expipe register-surgery SUBJECT --date 14.05.2017:12:35 --procedure implantation --weight 400
Drive adjustment¶
The first time you want to adjust the brain-area X, Y
(reffering to PAR.TEMPLATES['implantation']['X']) of a drive
an amount of um you need to
initialize an adjustment action with the command:
$ expipe adjust SUBJECT --init
When you have initialized and later make a new adjustment:
$ expipe adjust SUBJECT --date dd.mm.yyyyTHH:MM --adjustment X um --adjustment Y um
Note that in stead of --date dd.mm.yyyyTHH:MM can use -d now.
Working with Axona¶
Open Anaconda 3 prompt:
activate phy
- To get an overview of available commands::
- expipe –help
- For help på en spesifikk kommando::
- expipe axona register –help
To overwrite existing data use --overwrite:
expipe axona register z:\USER\DATA\SUBJECT\FILENAME.set
Now modules are loaded to the firebase server and depth registration is promted to the user.
Spikesorting:
$ expipe spikesort ACTION-ID
Manual spikesoring with phy:
$ phy neo-gui c:\user\uiousername\expipe_temp_storage\ACTION-ID\main.exdir --channel-group 0
Save with ctrl-s
Working with Open Ephys¶
Open Anaconda 3 prompt:
activate phy
- To get an overview of available commands::
- expipe –help
- For help på en spesifikk kommando::
- expipe openephys register –help
To overwrite existing data use --overwrite:
expipe openephys register c:\open_ephys\data\1704_2017-05-13_13-52-26_03
Now modules are loaded to the firebase server and depth registration is promted to the user.
Todo
probefile etc.
Data processing:
$ expipe openephys process ACTION-ID
Todo
Detail the processing: ground channels, cmr vs car, filtering, probe
Manual spikesoring with phy:
$ phy neo-gui c:\user\uiousername\expipe_temp_storage\ACTION-ID\main.exdir --channel-group 0
Save with ctrl-s
Plotting og analysis¶
Plotting with notebook:
$ expipe generate-notebook ACTION-ID --run
kernel -> restart and run all
Make png figurer:
$ expipe analyse ACTION-ID --spatial
If you only want to look at a particular channel group:
$ expipe analyse ACTION-ID --channel-group 0
Register cells and transfer data to norstore
When your are finished with the analysis you want to produce a representation of the channel groups on the firebase server. This can be performed with the command:
$ expipe register-units ACTION-ID --tag GC --tag BC --message "found a beatiful grid cell on channel group 2"
Example tags: Bare dritt: no, Good shit: yes, Vet ikke: maybe, Head direction: HD, Grid cell: GC, Place cell: PC, Spatial cell: SC, Boarder cell: BC.
Go to SERVER/PROJECT/ACTION-ID/main.exdir/analysis to check out the results.
Developers’ guide¶
Getting the source code¶
We use the Git version control system. The best way to contribute is through GitHub. You will first need a GitHub account, and you should then fork the repository.
Working on the documentation¶
The documentation is written in reStructuredText, using the Sphinx documentation system. To build the documentation:
$ cd expipe/docs
$ make html
Then open doc/build/html/index.html in your browser.
Committing your changes¶
Once you are happy with your changes, run the test suite again to check that you have not introduced any new bugs. Then you can commit them to your local repository:
$ git commit -m 'informative commit message'
If this is your first commit to the project, please add your name and
affiliation/employer to doc/source/authors.rst
You can then push your changes to your online repository on GitHub:
$ git push
Once you think your changes are ready to be included in the main Neo repository, open a pull request on GitHub (see https://help.github.com/articles/using-pull-requests).