About us
Espace utilisateur
INSTN offers more than 40 diplomas from operator level to post-graduate degree level. 30% of our students are international students.
Professionnal development
Professionnal development
Find a training course
INSTN delivers off-the-self or tailor-made training courses to support the operational excellence of your talents.
Human capital solutions
At INSTN, we are committed to providing our partners with the best human capital solutions to develop and deliver safe & sustainable projects.
Home   /   Thesis   /   Discovery of new chromogenic probes for toxic using Chemistry-Trained Machine Learning

Discovery of new chromogenic probes for toxic using Chemistry-Trained Machine Learning

Computer science and software Engineering sciences Materials and applications


Today national and international situation justify new researches on the colorimetric detection of toxic and polluting gases (referred to as analytes in the following). For the already known and studied compounds, improvement of the detection capabilities involves increasing contrast and selectivity. For potential new analytes, it is also important to prepare for rapid identification of specific chromogenic probes. The objectives of the thesis will be to discover new chromogenic probes by using computational chemistry.
First stage of the thesis: Training of the model (ML/AI) on available database. This part of the thesis will focus on establishing a precise and robust model to classify the large experimental database available from our laboratory's previous work. This involves correlating the colorimetric results with the structures and chemical properties of the molecules described by state-of-the-art methods (e.g., https://pubs.acs.org/doi/10.1021/acs.chemrev.1c00107). At the end of this learning process, we will have a predictor (SVM, LCA, PCA…) validated on our data.
Second stage: Use of the predictor model to screen in silico several hundred thousand candidate probe molecules from commercial chemical libraries (and others), correlated with their chemical structure and property descriptions as in the first stage. After this initial screening, DFT prediction of the chromogenic response will be used to refine the selection of the best candidate molecules.
Third stage: Definition and implementation of an experimental chemical testing campaign. A fast organic synthesis platform HTE (high throughput experimentation) based on the miniaturization and parallelization of chemical reactions to optimize the implementation of synthesis reactions and tests, will save considerable time, while significantly increasing the number of possible combinations. HTE also enables the synthesis of libraries of analogous compounds. Following these massive tests, a second version of the predictor will be trained and will lead to the discovery of a new generation of chromogenic molecules.


Département des Technologies des NanoMatériaux (LITEN)
Service des Technologies Durables pour le Cycle des matières (DRT)
Laboratoire Mesure Sécurisation Environnement
Université Grenoble Alpes
Top envelopegraduation-hatlicensebookuserusersmap-markercalendar-fullbubblecrossmenuarrow-down