Researchers Turn To Artificial Intelligence To Reach New, Previously Unachievable Cancer Outcomes
Artificial intelligence (AI) has been used to gain understanding into the biophysics of cancer. A machine learning platform predicted a trio of reagents capable of generating a one-of-a-kind cancer-like phenotype in tadpoles.
The study, conducted by researchers at Tufts University's School of Arts and Sciences, the Allen Discovery Center at Tufts, and the University of Maryland, Baltimore County, was published, Jan. 27, in the Journal Scientific Reports.
AI Model Successfully Predicted Unique Cancer-like Phenotype
The paper suggests how AI can be used to help human researchers in domains, such as regenerative medicine and oncology control complex biological systems, to gather previously unachieved results. The scientists had previously proven that pigment cells, called melanocytes, found in developing frogs could be transformed into a cancer-like, metastatic forms if their normal bioelectric and serotonergic signaling is disrupted. In that scientific research, the team used AI in the process of reverse-engineering a model to explain the complex process.
The researchers hypothesized that an all-or-none kind of coordination of cells across the tadpole body could be scientifically explained and controlled, after coming across what seemed as an odd result. In their extensive experiments, the scientists noticed that all the melanocytes in a single frog larva would either convert to the cancer-like form or they would remain completely normal. A third outcome, where only some of the pigment cells would convert in a single tadpole, never occurred.
As part of the current research, the scientists gave the AI model the task of finding a way to achieve partial melanocyte conversion in the same animal, using either one or more interventions.
"Computational methods can reverse-engineer mechanistic models of tissue patterning and shape formation from expression data and experimental phenotypes. The utility of these methods is their ability to find novel regulatory interactions and even novel necessary regulatory genes. These methods are indeed becoming indispensable for understanding the complex coordination of signals necessary to develop and maintain correct body shapes and organs," noted the study.
The AI Model Proposed A Three-Step Treatment
The AI predicted a treatment composed of three parts, which the researchers admitted they wouldn't have managed to achieve. The system, therefore, successfully completed the task it was given and led to an unprecedented result in years of experiments the researchers carried out.
"Remarkably, applying the predicted combination of three reagents in vivo revealed precisely the expected novel outcome, resulting in partial conversion of melanocytes within individuals. This work demonstrates the capability of automated analysis of dynamic models of signaling networks to discover novel phenotypes and predictively identify specific manipulations that can reach them," concluded the research.
This type of approach is a key moment for research in the field of regenerative medicine, as one of the major obstacles is managing to manipulate the complex networks to reach an intended therapeutic result. Biological regulation is highly complex, which makes it notably complicated to fully understand the mechanical functioning and to predict resultant experimental outcomes. This type of model allows researchers to systematically interrogate the system, which makes it much easier to come across the precise interventions needed for the targeted outcome.
After this very successful attempt, researchers hope to extend their approach to other areas of regenerative medicine, thus improving the efficiency of cancer treatments.