Researchers from universities in Singapore and China have developed MolMapNet, a new artificial intelligence tool.
Although AI tools are usually good forrecognition of spatially ordered images (for example, images of objects), they do not work well with disordered data such as molecular properties. This impairs their effectiveness in the analysis of pharmaceuticals. Scientists have sought to overcome this limitation. The goal is to improve the performance of deep learning models for predicting pharmaceutical properties of drugs.
Researchers' innovativeready-made artificial intelligence tool MolMapNet for deep learning in predicting pharmaceutical properties. Starting with a molecule (top right), its molecular properties (such as molecular components below the molecule) are projected onto a 2D plate (top plate of a multi-plate structure) as an image recognizing image AI (multi-plate structure) reads image pixels for recognition indicators of pharmaceutical properties, and then predicts (two layers of interconnected links under the multi-plate structure) pharmaceutical properties (drug and bottle in the lower left corner). The box that opens up (bottom right) indicates that the AI tool can be used by non-specialists out of the box. Credit: Shen et al.
The deep learning model was created in three stages.
- The first is a broad study of the internal relationships of molecular properties of more than 8 million molecules.
- The second is the use of a newly developeddata transformation techniques for displaying the molecular properties of pharmaceuticals in 2D images. Pixel layouts reflect the internal relationship between these properties. They contain important indicators of pharmaceutical properties that are captured using trained deep learning models.
- The third is to train the MolMapNet tool to recognize 2D images and use them to predict pharmaceutical properties.
As a result, AI can capture certain patterns.a layout of pixels that characterize certain pharmaceutical properties. It looks like that. how artificial intelligence distinguishes between men and women in the image, studying different gender characteristics.
Innovative AI does not require fine-tuning of parameters, which means it is accessible to non-specialists.
Read more
New uranium compound breaks record for anomalous conductivity
The most stormy place on Earth: why Drake Passage is the most dangerous route to Antarctica
Evolutionary error: which organs in the human body work illogically