Neuroscience Gateway - Modeling, Data Analysis
and Software Dissemination on Supercomputers
Satellite Workshop - 2024 Society for
Neuroscience Annual Meeting, Chicago, IL
Date and Time: Saturday,
October 5, 2024, 08:30 AM - 11:30 AM (US Central Time)
Location: Social
Sciences Research Building (Social Sciences Common Room (SSRB 201))
1126 East 59th
Street
University of Chicago
Chicago, IL 60637
Registration (is free but is required given the limited space, please register here): https://na.eventscloud.com/2024.10-nsg-workshop
Organizers: Amit Majumdar, Subhashini Sivagnanam, Kenneth
Yoshimoto,
San Diego Supercomputer Center, University of California San
Diego
Ted Carnevale,
Neuroscience Department, Yale School of Medicine, Yale University
Co-organizers: Kimberly Grasch and H. Birali Runesha, University of
Chicago
Workshop theme: The Neuroscience
Gateway (NSG) serves the neuroscience community by providing researchers and
students easy, open and free access to a large number of neuroscience software
and tools on supercomputing resources, academic cloud computing resources, and
associated storage resources, which are located at various national academic
supercomputer centers in the US. NSG can be used by neuroscientists from any
country. NSG eliminates administrative and technical barriers for all researchers
and students and enables neuroscientists to carry out research that requires
large scale modeling and data processing using data from neuroscience data
archives. NSG enables dissemination of neuroscience modeling and data
processing software to the community. This workshop will bring together researchers
and educators from the broader neuroscience community and they will share their
perspective on neuroscience research, education and broad outreach as it
relates to computing, data archives, data sharing, training and teaching.
AGENDA
8:30 AM - 8:45 AM
Welcome
8:45 AM - 9:15 AM
Title: Human Neocortical Neurosolver: A Software
Tool for Cell and Circuit Level Interpretation of MEG/EEG signals
Authors: Nicholas Tolley,
Stephanie Jones
Affiliation: Brown
University
Abstract: MEG/EEG signals are correlated with nearly all healthy
and pathological brain functions. However, it is still extremely difficult to
infer the underlying cellular and circuit level origins. This limits the
translation of MEG/EEG signals into novel principles of information processing,
or into new treatment modalities for pathologies. To address this limitation,
we built the Human Neocortical Neurosolver (HNN): an open-source software tool
to help researchers and clinicians without formal computational modeling or
coding experience interpret the neural origin of their human MEG/EEG data. The
foundation of HNN is a biophysically-detailed neocortical model, representing a
patch of neocortex receiving thalamic and corticocortical drive. The HNN model
can be accessed through a user-friendly interactive graphical user interface
(GUI) or through a Python scripting interface. Tutorials are provided to teach
users how to begin to study the cell and circuit level origin of sensory event
related potentials (ERPs) and low frequency rhythms. The package is available
to install with a single command on PyPI (pip install
hnn_core), is unit tested and extensively documented.
HNN is additionally accessible through computing resources offered by the
Neuroscience Gateway (NSG) enabling large simulation workloads. We will give an
overview of the background of HNN, describe the newest features added to the
software, and highlight recent research projects using HNN.
9:15 AM - 9:45 AM
Title: The NEMAR Neuromagnetic Data, Tools,
and Compute Resource
Authors: Scott Makeig1, Kenneth
Yoshimoto2, Choonhan Youn2, Dung Troung1,
Subhashini Sivagnanam2, Amitava Majumdar2, Arnaud Delorme1
Affiliation: 1Swartz Center for Computational
Neuroscience, 2San Diego Supercomputer Center, University of California San
Diego
Abstract:
The recent BRAIN Initiative, funded by the Obama administration,
propelled the creation of archives of publicly and other funded scientific data
of all types. For human functional neuroimaging, the OpenNeuro.org archive was
funded to collect and publicly share data of all types. Its creator, Russ Poldrack, is an fMRI expert. For other imaging modalities,
NIMH funded projects to curate data contributed to OpenNeuro.
Our NEMAR.org serves that purpose for 'neuroelectromagnetic' data (EEG, MEG, iEEG). Beyond simple data curation, publication of data
quality measures and data visualization, NEMAR exemplifies what I believe
should become the basic unit of open science, what I call the 'integrated data,
tools, and compute resource' (datcor). By teaming
with the Neuroscience Gateway team, NEMAR now supports users worldwide in
identifying and performing sophisticated computations on increasing amounts of
publicly available data - without need for at-best balky data downloads.
9:45
AM - 10:15AM
Title:
DANDI: Building a collaborative ecosystem for neuroscientific data
Authors: Satrajit
Ghosh
Affiliation: McGovern
Institute, MIT
Abstract: The DANDI
Archive is a community-oriented platform designed to support the sharing,
analysis, and re-use of neurophysiology and microscopy data, using the BRAIN Initiative supported
standards for data sharing. By providing an open, FAIR-compliant (Findable,
Accessible, Interoperable, Reusable) ecosystem, DANDI facilitates collaborative
research in neuroscience. It integrates diverse data types, from
electrophysiology to imaging, ensuring that researchers can contribute,
discover, access, visualize, and compute on standardized datasets seamlessly.
This presentation will highlight the key features of DANDI, including its data
management infrastructure, open-source tools, and how it promotes transparency,
reproducibility, and interdisciplinary collaboration in neuroscience research.
10:15 AM - 10:30 AM BREAK
10:30 AM - 11:00 AM
Title: Enhancing
Perspectives in Neuroscience Research through Diverse Institutional
Partnerships.
Author: Elba Serrano
Affiliation: New Mexico State University
Abstract: Diversity in collaboration is expected to
yield more inclusive and representative research outcomes and potentially
address more varied needs within a field.
By pooling knowledge and resources from different disciplines,
institutions, and investigators, researchers can approach problems in
neuroscience from multiple angles, leading to more comprehensive and innovative
solutions. The nation's 700+ federally designated minority serving institutions
(MSIs) comprise about 15% of all degree-granting institutions and educate over
5 million students. This presentation will introduce attendees to the rich constellation
of MSIs with a spotlight on Hispanic serving institutions (HSIs), where 65% of
the nation's Latino students seek degrees. Drawing on experiences as lead for
the NSF HSI National STEM Resource Hub, the speaker will provide an overview of
the benefits and challenges in developing collaborations with colleagues at
MSIs, as well as strategies for identifying partners and developing authentic
relationships that further neuroscience research.
11:00 AM - 11:30 AM
Title: Bio-realistic
modeling of the mouse primary visual cortex using large-scale datasets
Authors: Shinya Ito1, Darrell Haufler1,
Kael Dai1, Joe Aman1, Javier Galván
Frail2, Guozhang Chen3, Claudio
Mirasso2, Wolfgang Maass4, Anton Arkhipov1
Affiliation:
1. Allen Institute, Seattle, Washington, USA
2. IFISC, University deles Illes Balears, Palma de Mallorca, Spain
3. Peking University, Beijing, China
4. Graz University of Technology, Graz, Austria
Abstract: Accurate models of cortical circuits facilitate a
deeper understanding of how neural dynamics are shaped and maintained within
the brain. We have developed an enhanced, biologically realistic model of the
mouse primary visual cortex (V1), building on the framework established by Billeh et al. (Neuron, 2020). This updated model integrates
new synaptic physiology data from Campagnola, Seeman
et al. (Science, 2022; portal.brain-map.org/connectivity/synaptic-physiology)
and detailed connectomics from the IARPA MICrONS dataset (www.microns-explorer.org), refining its
connectivity and synaptic dynamics. The resulting model exhibits stable
activity patterns before optimization.
Moreover, we utilized TensorFlow-based optimization techniques
to align model parameters with physiological data, including Neuropixels recordings, achieving key empirical targets
such as firing rates and orientation selectivity. This improved model not only
enhances our understanding of cortical processing but will also be made
publicly available to support further research.