A group of researchers from the Massachusetts Institute of Technology (MIT) have developed a solution that could make a difference in biological research, especially in the field of machine learning called BioAutoMATED.
This development was carried out by a group of teams led directly by Jim Collins as a professor in the Department of Biological Engineering. They developed a machine learning system called BioAutoMATED. This system is expected to eliminate the need for extensive machine learning expertise.
Traditionally, building a machine learning model is a process that can take a lot of time, and most importantly requires the expertise of special researchers. However, the latest systems using AI technology can streamline this process by automatically selecting and compiling the most suitable model for a given data set.
Not only that, thanks to this it is able to handle tedious data processing tasks, reducing the processing time that initially could take months to become as short as just a few hours.
Basically the focus of biological sequencing systems with machine learning can be differentiated through text and image recognition. This biological sequence, consisting of DNA, RNA, proteins, glycans, has intrinsic standard properties similar to the alphabet making it ideal for applying machine learning techniques in relevant research.
Now, by integrating several tools into one platform, fortunately BioAutoMATED is able to expand the search space beyond the capabilities of individual tools.
So what is the function of this system?
BioAutoMATED offers a wide variety of models supervised machine learning which includes binary and multi-class, as well as regression models. This system will help determine the optimal amount of data needed to train the model effectively.
The impact of this system can accelerate the overall research process. This system can lower barriers to entry so that domain experts in biology can explore and run experts machine learning in particular.
Currently the results of these innovations have received support from various grants and organizations. Including funding from Defense Threat Reduction Agencyprogram Defense Advance Research Projects Agency SD2, Paul G. Allen Frontiers Groupand other sources have made possible the realization of this system and its integration into Antibiotics-AI Projects.
The hope is that researchers can develop a seamless future of biology practice using the fast-paced world of AI and machine learning.
In addition, this research has also been published in the Cell System journal entitled “BioAutoMATED: An en-to-end automated machine learning tool for explanation and design of biological sequences”.
2023-07-23 15:54:07
#BioAutoMATED #Biology #Research #Innovation #MIT #Scientists
