-
Notifications
You must be signed in to change notification settings - Fork 0
Home
![wisp logo](https://github.com/Tharos-ux/wisp/raw/master/preview/WISP.png)
This wiki aims to describe program structure in order to keep it maintainable. It will focus upon the python files and their core functions rather than bash scripts, usage or biological data treatment. For a quick start, user guide section will cover how to install and how to call for classes. For developers, you may find class structure here.
In this first section, I will focus on explaining how WISP works, not on a programmatic level but rather where the files are stored, which are the folders you may want to edit or not and such. The structure can be a little confusing, but it aims to accommodate at best with all limitations. You can set up all databases and reference genomes, outputs and inputs outside of the program (when you query large amounts of data, it is useful to do so) : the parameters files (which we will go more in-depth review later on) will give input and output paths for program.
For the following lines, I will assume you are inside the main wisp
folder. When you want to use any command described in the user guide, it is in this directory you must be in your terminal.
The folder named preview
contains some documentation and figures for this wiki and readme having figures and such. For parameters files, a folder parameters_files
containing files that were used for the article and the subsequent report. You may edit those, rename them, create new ones... according to your needs (see how to edit such files).
The folders wisp_lib
, wisp_view
and wisp_tools
contain mandatory functions to make the program work. Do not touch or edit those, unless you know what you're doing. I advise you don't store any file inside of them. Many .py
and .sh
are in root wisp
folder, same rule applies here!
By default, a data
folder will be created during the installation process. It aims to contain databases, though you can put them somewhere else if you need or want to (see parameters file for changing this). As of logging facilities, a logs
folder will contain all bug reports for all WISP's subprograms.
For reference genomes and unknown samples, a folder genomes
will be created by default, with inside train
and unk
sub-folders. The key takeaway here is to split your unknown sample and your reference data in two folders. You can name them however you like, and put them wherever you want as long as you edit accordingly the parameters file.
For results, a output
folder will be created. Same policy applies here.
If you use conda, and you built conda env with env.sh
as recommended (see Installation) you will get a wisp_env
folder inside wisp
.
WISP : Bacterial families identification from long reads, machine learning with XGBoost