In order to have a discovery through a Machine Learning algorithm, we need to have a large training data set. There was a problem in the prediction of molecular properties and the generation of new molecules. This can be solved by machine learning and deep learning approaches. But, to solve this through these approaches would require a large amount of training data.
The researchers’ goal is to accelerate the discovery of new drug molecules and the development of materials. To address this problem, MIT researchers have found a way to predict the molecular properties of a molecule using a small data set. The team of researchers created a machine learning model that automatically learns the language of molecules. It is known as ‘Molecular Grammar’. This technique is applied to a small data set, which is more convenient. It uses all the information or grammar of the small data set. Take the molecules with similar structures and understand the similarities between these molecules. The system understands the laws that govern the similarity of molecules through reinforcement learning. The accuracy and f1 score of the model is such that it comes closest to achieving its goal. Molecular grammar is broadly classified into two parts. The first part is called the metagrammar, while the second part is called the hierarchical approach.
This new Molecular Grammar system gave better results than several Machine Learning models. It gives better results with a very small data set compared to the data set used to predict molecular properties through machine learning models. It’s a powerful technique and can be applied to graph-based data sets as well. It makes it feasible for both regressions and classification approaches. Therefore, to push their research further, the research team sliced the training data set in half and found that this gave better results. This was one of the remarkable achievements they had.
This method finds its uses in several domains, such as the prediction of the physical properties of the glass transition temperature. The research team would like to apply their molecular grammar model to 3D molecules and polymers. The model based on molecular grammar leads to the discovery of new molecules and also to the prediction of their properties.
review the Paper and MIT article. Don’t forget to join our 26k+ ML SubReddit, discord channel, and electronic newsletter, where we share the latest AI research news, exciting AI projects, and more. If you have any questions about the article above or if we missed anything, feel free to email us at [email protected]
🚀 Check out over 800 AI tools at AI Tools Club
