Researchers from the Department of Chemical Engineering of the University Rovira I Virgili (URV) developed a probabilistic model that allows identifying equivalent neurons between different brains, reports the Spanish university.
The method, published in the scientific journal Nature Communications, part of the idea that all the networks that are aligned are copies with errors of the same underlying structure, such as a type of plane or “blueprint”, and reconstructs this common pattern from the available observations.
“He assumes that all connectomas – neuronal connections maps within the brain – are copies with error of the same underlying connections pattern,” explain the authors of the study.
“Our main motivation was to be able to compare connectomas. But, to compare them, you must first know what neurons are equivalent in each brain,” says the researcher of the Department of Chemical Engineering Marta Sales-Pardo.
You may be interested: study discovers a new form of brain learning based on trial and error
New method would represent a step for neuroscience
The method was validated with real worm connectomas data C. elegans In different stages of development, of the larva of the Drosophila melanogaster (known as vinegar fly) and also with social communication networks by email.
“In all three cases, his technique has demonstrated a much higher accuracy to the methods that exist so far. We have been able to align up to ten networks at the same time, and we have done it better than any other tool known so far,” says the professor of the Department of Chemical Engineering Roger Guimerà.
Although the work is framed in the context of neuroscience, the method is applicable in any network where equivalence patterns can be sought.
The team also tested it with emails networks: “We started from the knowledge of who each user was and this has allowed us to validate that the method correctly found who is who through the connections,” explains researcher Teresa Lázaro.
It could also be applied in networks of interactions between proteins – uses to identify unknown functions in new or pathogens – or to detect suspicious patterns in financial or safety networks.
With EFE information
Follow us on Google News to always keep you informed
