To better understand the relationships between interindividual variability in brain regions’ connectivity and behavioural phenotypes, we are developing a connectivity-based psychometric prediction framework (CBPP). Preliminary to the development of this region-wise machine learning approach, we performed an extensive assessment of the general connectivity-based psychometric prediction (CBPP) framework based on whole-brain connectivity information.
Because a systematic evaluation of different parameters was lacking from previous literature, we evaluated several approaches pertaining to the different steps of a CBPP study.
We hence tested 72 different approach combinations (3 types of preprocessing x 4 parcellation granularity x 2 connectivity methods x 3 regression methods = 72 combinations) in a cohort of over 900 healthy adults across 98 psychometric variables. Overall, our extensive evaluation combined to an innovative region-wise machine learning approach offer a framework that optimises both, prediction performance and neurobiological validity (and hence interpretability) for studying brain-behaviour relationships.
A SciUnit library for data-driven testing of basal ganglia models. Employed for testing via the HBP Validation Framework. This test shall take as input a BluePyOpt optimized output file, containing a hall_of_fame.json file specifying a collection of parameter sets. The validation test would then evaluate the model for all (or specified) parameter sets against various eFEL features.
The pipeline ingests data from multiple measurement types of spatially organized neuronal activity, such as ECoG or calcium imaging recordings. The pipeline returns statistical measures to quantify the dynamic wave-like activity patterns found in the data. Individual parts of the snakemake-based pipeline are fully configurable. The composition of Cobrawap elements can be adapted to various datasets through by means of a modular design of self-contained sequential stages composed of multiple atomic blocks.
The EBRAINS Collaboratory offers researchers and developers a secure environment to work with others. You control the level of collaboration by sharing your projects with specific users, teams or all of the Internet. Many researchers are sharing their work already; several services, tools, datasets, and other resources are publicly available, and many more are available for registered users.
EBRAINS is open and free. Sign up now for complete access to our tools and services.