Pathogens are similar to bacteria as they are clever enough to adopt and this makes hard to fight them off with antibiotics. However, Los Alamos National Laboratory scientists have come up with a new way to tackle the problem. They are using machine learning to combat pathogens.
The research has been published in Communications Chemistry journal. It mainly focuses on identifying certain molecular properties and this could help in discovering new antibiotics. It is learned that the new antibiotics may be helpful against bacteria that are becoming resistant to current drugs.
Los Alamos National Laboratory scientist Gnana Gnanakaran said that some bacteria are good at resisting antibiotics and finding compounds that can penetrate and stop these bacteria is very tough. Their approach helps in digging into the molecular details of bacteria and this is something very crucial in developing successful drugs.
It is true that bacterial defenses are tough. Gram-negative bacteria have outer layer and it is hard for antibiotics to get through it. Moreover, they are capable in pushing out compounds that manage to get inside. This makes antibiotics less effective.
The research team is taking the help of machine learning to overcome the issue. They have developed a model and claims it could pinpoint which properties of certain compounds would help them penetrate the defenses as well as stay inside.
The study mainly focused on bacteria called Pseudomonas aeruginosa, which is very common in infections. They analyzed more than a thousand different compounds with the help of machine learning and learned how they interacted with the outer layer of the bacteria.
The findings highlight what properties make a compound effective against Pseudomonas aeruginosa and also paves the path for similar studies on other bacteria.
