GSON module "chimie informatique sous python"

This course is mostly based on teachopencadd.
Huge thanks to them for providing such good learning material.

Launch notebooks directly in your browser

• If you prefer the modern jupyter lab interface:

  • be careful that for the exos using nbautoeval, the cell output may return error
  • for the moment it is NOT recommended to use above method

• If you prefer the traditional jupyter notebook interface:

The materials are exactly the same. The only difference is the appearance.
The 1st launch can take around 10 minutes, and it will be faster from 2nd time.
It does not require any setup/installation on your machine (which is quite easy to use). You just need an Internet connection and a browser.

Remarks:

1. To save resources, your connection to the Binder server will be automatically cut off if you have no activity (cell edit/cell execution) for around 10 mins

2. Your changes to the notebooks will NOT be saved. So if you close the browser and re-open Binder, you lose all what you have done previously...

Special instructions for promo 2023-2024

If you are using Windows on your own PC, before each course, please execute the command git pull in your terminal, to get the latest version of material.

If you are using the Linux session on the PCs of "salle info", here are the commands to follow, from Tuesday to Friday:

1. Open a terminal, copy-paste below command into it, and execute it:

/opt/anaconda3/bin/conda init

2. Normally, the message of the terminal will suggest you to close and re-open the terminal after above command is executed successfully.
   Simply close and re-open another terminal.
   You should now see (base) on the leftmost of your prompt. (ask the teachers if you still cannot see it)

3. Create the virtual environment for this course from given file, by copy-paste below command into your terminal, and execute it:

conda env create -f https://raw.githubusercontent.com/ICOA-SBC/GSON-cheminformatics/master/environment.yml

• This file contains a list of conda/pip packages that are required for this course.
• You are encouraged to check the content of this file, to make sure it does not contain malicious software.
• This command will take 3-10 mins, depending on the configuration of your PC, and the Internet connection

4. Once above command finishes with success, you can now clone the repo to your local PC, by copy-paste below command into your terminal, and execute it:

git clone https://github.com/ICOA-SBC/GSON-cheminformatics

5. Activate the virtual environment:

conda activate teachopencadd

• You should observe that the (base) has become (teachopencadd), meaning that you are now in this new virtual environment.

6. You are ready to go. Simply launch the notebook interface with command:

jupyter notebook

Organisation of session 2023-2024

Lundi 15 janvier 2024 13h30-17h30

• Introduction
  • Slides 1_GSON_intro (30 min) XM & JM
  • Slides 2_introduction_informatique (45 min) XM
  • Talktorial Introductif 1 Intro Python, jupyter (1h40) XM

Mardi 16 janvier 2024 13h30-17h30

• Talktorial Introductif 2 Intro chemoinfo - RDKit (1h40) JM

• Data Acquisition (1h30) XM

  • Talktorial 1 Data acquisition from ChEMBL

Mercredi 17 janvier 2024 13h30-17h30

• Filtering (1h45-2h) JM

  • Descriptors and ADME (1h30)
    • Slides 4_Descripteurs_fingerprints
    • Talktorial 2 Molecular filtering: ADME/Lipinski criteria

• Filtering (40 min) XM

  • Talktorial 3 Substructure removal : PAINS

Jeudi 18 janvier 2024 13h30-17h30

• Ligand based Screening (2h30) JM

  • Slides 5_fingerprint_similarity (30 min)
  • Talktorial 4 Fingerprints and Molecular Similarity (2h)

• Clustering (1h45) XM

  • Talktorial 5 : Compound clustering

Vendredi 19 janvier 2024 13h30-17h30

• Machine Learning (1h45) JM

  • Slides 6_Machine Learning
  • Talktorial 7 Machine Learning (ROC curve)

• Applications in chemoinformatic (45 min) XM & JM

  • Slides 7_Applications

• Exam (60 min)

History of previous years

• 2018: Fabrice, Colin

• 2019: Colin, Gautier

• 2020: Gautier, Pierre-Yves

  • 21 students
  • Integration of TeachOpenCADD and binder

• 2021: Gautier, Pierre-Yves

  • 22 students
  • Distance learning, via TEAMS
  • Integration of nbautoeval (Exercises and Quiz)

• 2022: Pierre-Yves, Xiaojun

  • 9 students (7 present, 2 absent)
  • Introduce mamba for quicker env resolution
  • Added short introduction of Linux Commands

• 2023: Pierre-Yves, Xiaojun

  • 5 students (4 present, 1 absent)
  • Created a pre-course questionnaire to better adapt the course to future students attending the course

• 2024: Xiaojun, Jérémy

  • 17 students (16 inscriptions + 1 volunteer participant)
  • Cleaned not used packages/channels in environment.yml for quicker env resolution
  • Added Slide 7_Applications, to showcase other fields of applications not covered in the course

Installation

Suppose that you are using Linux (MacOS should work the same way, since it is also Unix):

1. Install miniconda (or anaconda if you prefer)

• For installation details, please refer to their official documentation.

2. Open a terminal, you should normally see (base) before your prompt. (See below FAQ part if you cannot see (base))
   Create the virtual environment for this course from given file:

conda env create -f https://raw.githubusercontent.com/ICOA-SBC/GSON-cheminformatics/master/environment.yml

• This file contains a list of conda/pip packages that are required for this course.
• You are encouraged to check the content of this file, to make sure it does not contain malicious software.

3. Open a terminal, and clone the repo to your local PC:

git clone https://github.com/ICOA-SBC/GSON-cheminformatics

4. Activate the virtual environment:

conda activate teachopencadd

• You should observe that the (base) has become (teachopencadd), meaning that you are now in this new virtual environment.

5. You are ready to go. Simply launch the notebook interface with command:

jupyter notebook

FAQ

What if you cannot find conda by typing which conda, or you do not see base before your prompt?

Experience 2024:
In the computer rooms of COST, conda is installed at /opt/anaconda3, not under each user's /home.
Furthermore, it is NOT added to the $PATH, making it unfindable by which conda command.
Even more, the /home of each student will be automatically deleted on the 2nd day of the course, after restarting the PC...(by default the PCs are automatically shut down during the night).

So it means the students have to repeat the steps of

• initiate conda
• create conda virtual environment
• clone the course material

from Tuesday to Friday, before each course starts (which is annoying!)

Work in progress, reserved for teachers

1. Open a terminal, and type echo $PATH to check the existing PATHS (normally you shoud not be able to find /opt/anaconda3).
2. Type gedit ~/.bashrc. A text editor interface will automatically appear.
3. Add export PATH=/opt/anaconda3/bin:$PATH to the end of the file, then save and close the file.
4. Reload the profile with source ~/.bashrc
5. Type the command which conda, you should be able to find it at opt/anaconda3/.
6. Type the command conda init bash to initiate it.
7. Reload the profile with source ~/.bashrc.
   After all these steps, you should now see (base) before your prompt.

Another solution provided by service info (which seems easier, thus recommended)

1. Open a terminal, and change to the directory where conda is installed: cd /opt/anaconda3/bin
2. Initiate it by ./conda init
3. Close the terminal, and re-open a new one
4. You should now see (base) before your prompt

A more automatic way (WIP, since it requires restarting the terminal, it does not work for the moment)

An init.sh script is also included in the repo. It automates all the 3 major steps:

• export conda to $PATH
• create environment
• clone the repo

To use it under Linux:

• Download this file, and put it at your /home
• Open a terminal, and execute the downloaded script by bash init.sh
• This process will take ~3 mins (the resolution of environment may take extra time)
• If the script success, you will need to close and open another terminal

Voilà! You are ready to go!