English (United Kingdom)

https://curated-unify.zendy.io/wp-json/zendy-region/v1/featured_content/oa?rat=en

https://curated-unify.zendy.io/wp-json/zendy-region/v1/highlighted_journal/

Global: Zendy Plus annual

Presents the access of premium content as premium feature

Premium Content

Presents the keyphrase highlighting as premium feature

Keyphrase Highlighting

Presents the summarisation as premium feature

Summarisation

Insights

Presents the pdf analysis as premium feature

PDF Analysis

Presents the zaia usage as premium feature

ZAIA

Global: Zendy Plus monthly

Global: Zendy Tools annual

Global: Zendy Tools monthly

Global: Zendy Free Plan

Kohonen neural network or SOM (Self-Organizing Maps) and k-means are effective computational techniques for extracting information from a large dataset, in some cases, more suitable than traditional statistical methods (such as regression) for detecting multivariate data patterns. We used these two computational approaches to study the developmental biology of a large dataset of weight versus age in common marmosets, seeking to provide a more precise description of the onset and end of behaviors during ontogenesis. Both techniques perform clustering, a technique that falls into a group of undirected data mining tools. The goal of undirected data mining is to discover structure in the data as a whole, separating or discovering classes or groups that are not easily detected from direct data inspection. Both k-means algorithms and self-organizing maps are clustering methods that classify patterns without the need of previous information on data distribution, as occurs in supervised methods. When used in feature datasets (for instance, age and weight), these techniques allow a data mining process that identifies stages and substages of the data distribution being used. In particular, the SOM algorithm is able to map features from a highdimensional space into a oneor two(in fact any dimension less than the dimension of the data) dimensional space with preservation of topology. Thus, these tools are better for modeling and identifying stages in a sample pool and they enable us to visualize data distribution in a two-dimensional space, even when the number of features in the data is high (or the topology is complex). For a same database we found that both approaches converged to a same result, enabling the process of classification, thus achieving the goal of this study. Biomedical research using the common marmoset (Callithrix jacchus) continues to grow, through the investigation of different topics such as infectious diseases, neuroscience and development, toxicology and drug development, reproductive biology and behavior. Using this classification we found that the infantile stage of the current classification could be divided into three stages and that feeding behavior is statistically different among them. We also applied a statistical method, Principal Component Analysis (PCA) to evidence the sex difference. Using data collected from 9 animals (4 males and 5 females) during development evidenced a sex difference, where male behavior features are more dispersing than those of females, indicating that behavior is more changeable in males. These findings indicate that the use of clustering techniques provides a more precise description of changes during ontogenesis in common marmosets

Computational Approaches as a Tool to Study Developmental Biology in New World Primates