BOLD signal reconstruction and simulation from cellular data-driven models


Project Summary

Functional signals directed by magnetic resonance imaging (fMRI) techniques are widely used in neuroscience to monitor brain activity noninvasively. The overarching scientific issue is to connect these signals to the underlying activity of neurons. We will therefore generate a workflow able to reconstruct the blood oxygen level dependent (BOLD) signal starting from cellular-level models available in the Brain Simulation Platform (BSP). This workflow will be crucial to simulate the BOLD fMRI maps in the forthcoming Human Brain Project (HBP) European modeling ifnrastructure (EBRAINS), especially by using the whole-brain simulator "The Virtual Brain" (TVB).

The BOLDsim project is based on detailed neuronal microcircuit models generated in HBP during SGA1 and SGA2 and currently available on the BSP. BOLDsim is intended to push forward brain signals reconstruction at the local circuit level and to apply this reconstruction to whole brain simulations. The final aim is to generate BOLD fMRI maps in TVB/EBRAINS.

Collaboration with the HBP

In this project we will reconstruct the BOLD signal associated to patterns of activity (rs-FMRI, td-fMRI), provide an effective pipeline in EBRAINS applicable to the cerebellar microcircuit model and extendable to other brain structures (eg. the cerebral cortex), and provide a proof of principle of this procedure for brain signal anylsis in TVB.

The work will be based on the different branches of EBRAINS and will itself provide new models and workflows in pyNEST and pyNEURON to be uploaded and made publicly available. Conversion into ARBOR will be considered. 

The Brain Simulation Platform (BSP) will be used to import the scaffold of the cerebellum model, providing the differentiated neural populations and their spatial positioning, as well as the connectivity algorithms (following distance-based and geometry-based rules). The new BOLDsim cerebellum circuit will be integrated in the TVB through pyNEST codes.

BOLDsim will release the first bottom-up simulation of BOLD signals using realistic models in the cerebellar network, applicable to the whole brain, intergated into EBRAINS and TVB.

Both experimentalists, modelers, and theoreticians will benefit from BOLDsim.

Partnering Organisations


Project Coordinator

Shyam Diwakar is the Director of Computational Neuroscience and Neurophysiology Laboratory, a Faculty fellow at the Amrita Center for International Programs and an Associate Professor at the School of Biotechnology, Amrita Vishwa Vidyapeetham (Amrita University), India. He is the Institute Coordinator and Co-investigator of Virtual and Accessible Laboratories for Universalizing Education (VALUE), a major virtual labs initiative supported by Sakshat mission of MHRD, Government of India, and Principal Investigator of few other projects funded by Department of Science and Technology.

Related publications

1. Nutakki C, Radhakrishna S, Nair B, Diwakar S. Modeling fMRI BOLD signals and temporal mismatches in the cerebellar cortex, CSI Transactions on ICT, 2019, Springer Nature. 

2. Parasuram H, Nair B, Naldi G, D’Angelo E, Diwakar S, Understanding Cerebellum Granular Layer Network Computations through Mathematical Reconstructions of Evoked Local Field Potentials. Ann Neurosci 2018; 25:11-24

3. Parasuram H, Nair B, D‘Angelo E, Hines M, Naldi G and Diwakar S (2016). Computational Modeling of Single Neuron Extracellular Electric Potentials and Network Local Field Potentials using LFPsim. Front. Comput. Neurosci. 10:65. doi: 10.3389/fncom.2016.00065

Key facts

Time frame: 2020-2021