QuChemPedIA : Quantum Chemistry encycloPed and Intelligence Artificielle.
Invitation code : 3VwMu3-eTCg32
Molecular chemistry is lagging behind in term of open science. Although modelization by quantum mechanics applied to chemistry has become almost mandatory in any major publication, computational raw data is most of the time kept in the labs or destroyed. Furthermore, the software used in this area tend to lack effective quality control and computational details are usually incomplete in the articles and the information may not be reused or reproduced. The first objective of this project is to constitute a large collaborative open platform that will solve and store quantum molecular chemistry results. Original output files will be available to be reused to tackle new chemical studies for different applications. Machine learning and more generally artificial intelligence applied to chemistry data promises to revolutionize this area in the near future, but these methods require a lot of data that this project will be able to provide.
Today, it is impossible for a human to take into account the results, even limited to the most important data, for millions of known molecules. The second objective of this project is to radically change the approach developing artificial intelligence and optimization methods in order to explore efficiently the highly combinatorial molecular space. Generative models aim to provide an artificial assistant, which on the one hand has learned to predict the characteristics of a molecule and estimate its cost of synthesis, and on the other hand is able to browse effectively the molecular space. Generative models would open many perspectives by greatly facilitating the screening of new molecules with many potential applications (energy, medicine, materials, etc.). The bottleneck for our AIs is the computing power needed to verify the properties of the generated molecules.
By supporting this project, you will help chemical researchers around the world by building a unique collection of results. You will also help our AIs to propose much more new targets for the different applications we are addressing than we could do on our own.
Thank you for your help !
Thomas Cauchy (chemist)
Benoit Da Mota (computer scientist)
Already joined? Log in.
Big server failure
The server has been offline for 5 days due to a failure on the system disks. With a lot of work, we managed to get the server back online without any data loss. The disk redundancy is back online.
Sorry for the lack of news. The current campaign is still ongoing and we are also working on scientific publications. The health situation gives us a lot of extra work, but we don't give up!
17 Mar 2021, 14:51:21 UTC · Discuss
Scientific publication and news
Thank you very much for your help.
I am pleased to announce the publication of our latest open access article describing EvoMol, our opensource molecule generator.
EvoMol: a flexible and interpretable evolutionary algorithm for unbiased de novo molecular generation
The objective of this work is to design a molecular generator capable of exploring known as well as unfamiliar areas of the chemical space.
Our method must be flexible to adapt to very different problems. Therefore, it has to be able to work with or without the influence of prior data and knowledge. Moreover, regardless of the success, it should be as interpretable as possible to allow for diagnosis and improvement.
We propose here a new open source generation method using an evolutionary algorithm to sequentially build molecular graphs. It is independent of starting data and can generate totally unseen compounds. To be able to search a large part of the chemical space, we define an original set of 7 generic mutations close to the atomic level.
Our method achieves excellent performances and even records on the QED, penalised logP, SAscore, CLscore as well as the set of goal-directed functions defined in GuacaMol. To demonstrate its flexibility, we tackle a very different objective issued from the organic molecular materials domain. We show that EvoMol can generate sets of optimised molecules having high energy HOMO or low energy LUMO, starting only from methane. We can also set constraints on a synthesizability score and structural features. Finally, the interpretability of EvoMol allows for the visualisation of its exploration process as a chemically relevant tree.
You can find for free the full article :
or here :
You can help us a little bit more by visiting these pages to give us visibility and you can also share on your teams forums and/or social websites like tweeter.(@b_damota)
We have been working for some time now on the following article. It will deal in particular with the calculations you have made since the beginning of the project. The result will be an open access dataset. While we are writing this article we are also working on the next parts. Without divulging, I can only tell you that your calculations will help us to propose a unique tool that is particularly useful for chemists. We will of course keep you informed! Currently, two campaigns are in progress and do not concern the work mentioned above. The "nwchem long" units and the new "nwchem" with tasks with prefix "CL9" will also bring new results for articles probably end of 2021 or 2022.
29 Sep 2020, 8:09:19 UTC · Discuss
Scientific publication and new WUs
First of all, thank you very much for your help and for your interest in our research.
We are proud to announce the imminent publication of our work on the generation of molecules with AI. You can already read the first draft here : https://www.researchsquare.com/article/rs-36676/v1. It's a raw version with almost no formatting. We have a more polished version that should come out in few weeks. The article will be in open access, the molecule generator in open source and the data in open data. Open Science!
Our exploration of chemical space continues and we have just generated more than 2.5 million small molecules. I know that some people are waiting eagerly for the return of the short WU and here they are! As before, many calculations will be considered invalid because of unstable molecules, but this is the price for unbiased cartography of the chemical space. The first results are very encouraging and we hope that these 2.5 million new molecules will help to provide extremely useful tools for many chemists.
18 Jul 2020, 9:31:44 UTC · Discuss
NWChem long no longer in beta soon
The latest tests for long units (a few dozen hours) are conclusive with a good success rate. I don't plan to put this application in a VM for Windows and Mac, short units are already enough of a problem. The "NWChem Long" application will therefore come out of its beta phase soon. If you don't want to get these units that can last really long, don't forget to update your preferences.
Thank you for your help and your patience which allowed us to find good parameters.
24 Feb 2020, 17:43:47 UTC · Discuss
Thank you very much ! You are a really great help.
Initially, only two hundred "NWchem long" workunits should have been submitted for testing. Your numerous results and also the numerous failures allow us to better evaluate parameters for this new simulation. I still have to write some scripts for the new inputs, but I should be able to submit jobs soon. Don't hesitate to cancel in progress workunits (NWchem long).
18 Feb 2020, 9:01:04 UTC · Discuss
©2021 Benoit DA MOTA - LERIA, University of Angers, France