SkillfulAi

AI system can generate novel proteins that meet structural design targets

MIT researchers have developed machine-learning algorithms that can generate proteins with specific structural features using generative models. These proteins can be used to make materials with certain mechanical properties such as stiffness or elasticity. This type of biologically inspired material could replace those made from petroleum or ceramics with a smaller carbon footprint. The models can also be used to develop protein-inspired food coatings that can keep produce fresh longer and are safe for humans to eat. The models can quickly generate millions of proteins, giving scientists a portfolio of new ideas to explore. The researchers used attention-based diffusion models to build two machine-learning models that can predict a variety of new amino acid sequences that form proteins that meet structural design targets. The models operate at the overall structural properties of the protein and at the amino acid level.

AI mobile camera van catches 239 law-breaking motorists
Published

Safer Roads Humber, a partnership between police and councils, is using artificial intelligence (AI) to catch drivers in the act of using their phones while on the road. Equipped with multiple cameras, the technology captures footage which is then processed by AI before being reviewed by an officer. This innovative system has caught 56 people in South Cave, 73 in Hull, 55 in Willerby and 55 in Grimsby. Any driver detected using a phone while driving risks a £200 fine and six points on their driving licence. Those who fail to wear a seatbelt in a vehicle can also face a £100 fine, with drivers being held responsible for any passengers under 14 years of age. Safer Roads Humber has stated that some drivers may be given the option to attend an educational course as an alternative to prosecution.

when you think you can do without AI but in reality

How artificial intelligence is matching drugs to patients

Israeli health-tech firm Genetika+ is using stem cell technology and artificial intelligence (AI) software to match antidepressants to patients, avoiding unwanted side effects and ensuring the prescribed drug works as well as possible. From a patient's blood sample, Genetika+ can generate brain cells that are then exposed to several antidepressants, and recorded for cellular changes called "biomarkers". This information, combined with a patient's medical history and genetic data, is processed by an AI system to determine the best drug for a doctor to prescribe and the dosage. The technology is still in development, but Genetika+ plans to launch commercially next year. The company has secured funding from the European Union's European Research Council and European Innovation Council and is also working with pharmaceutical firms to develop new precision drugs. There are more than 280 million people globally who suffer from depression, according to the World Health Organization.