a02l01 / infest

Python script to compute kinematics of lesion development

INFEST

A new version of INFEST is proposed by Darcy Jones and is freely available

Overview

INFEST for kINematic oF lESion developmenT. This plugin was used to compute the kinematic of lesion caused by the necrotrophic fungus Sclerotinia sclerotiorum. INFEST was developed for QIP (quantitative immunity in plant) @ LIPM (Lab of plant microbes interaction) in Toulouse by Adelin Barbacci. INFEST was founded by Sylvain Raffaele's ERC varywim. Feel free to use it.

For academics use please cite :

Barbacci, A., Navaud, O., Mbengue, M., Barascud, M., Godiard, L., Khafif, M., Lacaze, A., Raffaele, S., 2020 Rapid identification of an Arabidopsis NLR gene conferring susceptibility to Sclerotinia sclerotiorum using real-time automated phenotyping. Rev.

We are on twitter AB, SR

INFEST

usage: infest.py [-h] [-f FIRST] [-l LAST] mpath

positional arguments:
  mpath                 Path to the directory containing pictures

optional arguments:
  -h, --help            show this help message and exit
  -f FIRST, --first FIRST
                        Number of the first picture
  -l LAST, --last LAST  Number of the last picture

output

analyse.txt file created in the mpath directory containing 3 columns: the Id of leaf, the time extracted from pictures name and the size of the Lesion

Examples

python infest.py mpath -f 0 -l 400

python infest.py mpath -f 0

python infest.py mpath

fit INFEST

usage: fit_INFEST.py [-h] [-ft FIRST] [-g] path_in path_out

positional arguments:
  path_in               the path to the file containing temporal data computed
                        by INFEST
  path_out              the path to the file containing LDT and Latency

optional arguments:
  -h, --help            show this help message and exit
  -ft FIRST, --first FIRST
                        the first time to consider for the computation of the
                        LDT
  -g, --graph           monitoring the fit of the curve

output

txt file specified in path_out directory containing 9 columns: the Id of leaf, the parameters a1 to a5 resulting from the fit, the residuals of the fit, the lesion doubling time LDT, and the Latency

Latest news

• Version 1 available

---

Getting started

Details

This version has been developed and tested under ubuntu 18.10 with python 2.7. The dependancies are:

• backports-functools-lru-cache==1.5
• cloudpickle==1.2.2
• cycler==0.10.0
• decorator==4.4.0
• kiwisolver==1.1.0
• matplotlib==2.2.4
• networkx==2.2
• numpy==1.16.5
• pillow==6.2.1
• pyparsing==2.4.2
• python-dateutil==2.8.0
• pytz==2019.3
• pywavelets==1.0.3
• scikit-image==0.14.5
• scipy==1.2.2
• six==1.12.0
• subprocess32==3.5.4
• argparse==1.1
• pandas==0.23.3

Dependancies are listed in the env_Infest.yml file

Prerequists

Python and conda

Install conda

• For linux install please see: here
• For other systems: here

Pictures & Files

• Jpeg images stored in a directory and named by an integer e.g. 1.jpg to N.jpg corresponding to the time course.
• the layout file grid_layout.layout in the subdirectory grid_layout of the directory containing pictures (e.g. my_pictures/grid_layout/grid_layout.layout c.f. tutorial)
• The layout file provide the Id and the bounding boxes of leaves e.g.

id_leaf_1\tymin\txmin\tymax\txmax\n
id_leaf_1\tymin\txmin\tymax\txmax\n
id_leaf_3\t...

with \t a tabulation.

Download INFEST and fit_INFEST

• manually, using git $ git clone https://github.com/A02l01/INFEST.git

Creation of the conda environment

Creation of a conda environment called INFEST from the yaml file . In bash

• Check that env_Infest.yml file is in your working directory
• $ conda env create -n INFEST -f env_Infest.yml
• Initialize your shell $ conda init bash
• To activate INFEST env: $ conda activate INFEST
• To deactivate INFEST env: $ conda deactivate

---

Tutorial

This tutorial was designed for linux users. It is easily transposable for macOS and Windows users by replacing most of command lines by fastidious mouse clicks.

In this short tutorial we will use INFEST to compute the kinematic of lesion development of a single detached leaf of Arabidopsis thaliana coined Col-0_154.

Download data

Data are in the data_tuto/ directory. Download and extract data with git:

$ git clone https://github.com/A02l01/tuto.git

Other kinematics can be computed by adding bounding boxes of leaves in the grid_layout.layout file.

creation of the layout file

• Downloaded data contains yet a layout file but in the general case you must generate this file and put in the right directory
• If needed creates a directory in the pictures directory and the file grid_layout.layout in grid_layout/ e.g. ./tuto/data_tuto/pictures/grid_layout/grid_layout.layout. Fill grid_layout.layout with the coordinates of the bounding rectangles of leaves. We used ImageJ to obtain coordinates. In this example the single line added in the grid_layout.layout file is

Col-0_154 1410 2574 1497 2721

with:

• Col-0_154 the Id of the leaf,
• 1410 ymin of the rectangle,
• 2574 xmin of the rectangle,
• 1497 ymax of the rectangle,
• 2721 xmax of the rectangle.

Items are separeted by a tabulation.

Compute kinematics of Lesion

• In the directory containing infest. py file, activate conda environment (to create the INFEST conda environment please see instructions here)

$ conda activate INFEST

$ python infest.py '../tuto/data_tuto/pictures/' -f 0 -l 270

results are stored in '../tuto/data_tuto/pictures/analyse.txt'

Compute the lesion LDT

We show that LDT is a good proxy of the level of plant resistance in Barbacci et al. 2020. Nevertheless other proxy could be derived from the kinematics computed by INFEST.

To extract the lesion doubling time (LDT) from the kinematic of lesion development using the python script fit_INFEST.py:

$ python fit_INFEST.py '../tuto/data_tuto/pictures/analyse.txt' '../tuto/data_tuto/pictures/ldt.txt' -g -ft 400 leading to

---

  ##     ###             ###     ##
 #       ###             ###       #
#                                   #
#        ###     ###     ###        #
#        ###     ###     ###        #
 #        #       #       #        #
  ##     #       #       #       ##

QiP Team LIPM Toulouse