91滴滴

Four projects have been awarded funding through the 2021 Neuro Community Open Science Initiative call supported by the Tanenbaum Open Science Institute (TOSI). Each of these funded projects facilitates the open sharing of scientific outputs with key stakeholders, including other researchers, patients, policymakers, and the public.

• The Neurodegenerative Disease Knowledge Portal

Sali Farhan (The Neuro), Mike Nalls (NIH - Center for Alzheimer鈥檚 and Related Dementias), Jason Flannick (MIT and Harvard)

• Video killed the written protocol: NeuroEDDU how-to vids

The Neuro Early Drug Discovery Unit (EDDU) team: Luisa Pimentel, Thomas Durcan, Paula L茅pine, Meghna Mathur, Cecilia Rocha, and Sophia Penner

• Open Science Office Hours

Kendra Oudyk (The Neuro)

• Psychosis SpeechBank

Lena Palaniyappan (Douglas Mental Health University Institute)

To learn more about each of these Open Science initiatives, please see below.

Stay tuned for the next Neuro Community Open Science Funding Call announcement. Coming Soon!

The Neurodegenerative Disease Knowledge Portal: An open-access genomics browser for neurodegenerative diseases

Summary

The Amyotrophic Lateral Sclerosis Knowledge Portal (ALSKP.org) is an open-access genomics browser composed of nearly 4000 ALS cases and 8000 ethnically matched controls that have been used by the community to access genomics data from ALS patients without the need for large, expensive computing resources. In this project, we plan to continue to expand the ALSKP by building a larger Neurodegenerative Disease Knowledge Portal (NDKP) that contains genomics data from multiple neurodegenerative disease datasets such as Alzheimer鈥檚 disease (AD), Parkinson鈥檚 disease (PD), frontotemporal dementia (FTD), and other allied neurological conditions. Users will be able to freely access genomic variants unique or common to all neurodegenerative diseases, as well as see their association metrics in each disease (e.g. significantly associated with a neurodegenerative disease or not.

Video Killed the Written Protocol: The NeuroEDDU how-to-videos for working with iPSCs

Summary

At the Early Drug Discovery Unit (EDDU), we are active participants in fostering and practicing Open Science. As a result of our efforts, we have generated and published online a collection of 20+ openly available ranging from handling induced pluripotent stem cells (iPSCs) to CRISPR editing, organoid generation and data analysis field. These protocols can be used by researchers and trainees across the globe and are openly available without restriction. These protocols, available on the Zenodo platform, have been viewed over 7000 times and downloaded more than 5,000 times within the past two years, ensuring what we do is getting into the hands of those that need it.

The pandemic has led researchers to become more visual, while at the same time putting technology at our fingertips to make our work more accessible. We propose to expand our portfolio of protocols by developing visual training guides. Our goal is a multi-language video collection鈥攗p to five of our written protocols will be converted to video format with closed captions. These videos will be available in a total of eight different languages for free and open access through our data portal, Vimeo channel and The Neuro鈥檚 YouTube channel. This material will be distributed through a Creative Commons license.

Open Science Office Hours (OSOH)

Summary

The goal of Open Science Office Hours (OSOH) is to empower students and researchers to make their work more open. We plan to do this by sharing Open Science resources, facilitating discussion, and offering individual assistance. We will start by engaging the neuroscience community at 91滴滴, and perhaps increase the scope in the future.

Many students learn Open Science skills at events like summer schools and hackathons, but might not have in-lab support for using what they have learned in their everyday research. Even principal investigators may need extra support to transition to a more open workflow. In 2020, we did a survey to assess the need for OSOH, and over half of our respondents listed 鈥渓imited support鈥� and 鈥渓imited skills鈥� as barriers to doing Open Science. OSOH aims to remove these barriers at the Neuro.

There will be two organizers consistently involved throughout the year. Each month will have a theme, such as 鈥淥pen data鈥� or 鈥淥pen hardware鈥�. At the beginning of the month, we will share resources on the theme, and people can sign up for help during office hours. At the end of the month, we will hold a live virtual discussion on the theme.

Psychosis SpeechBank: A collaboration for clinical linguistic research in psychiatry

Summary

Many adolescent onset mental disorders have a chronic intermittent course, with frequent and disabling relapses. We urgently need reproducible markers to track the illness course and enable wider implementation of early intervention at every stage of disorders such as psychosis. The form and content of speech provide the primary diagnostic and prognostic information for psychiatric practice. Our speech tracks our mental state; it remains the most accessible, remotely generated, inexpensively acquired, objectively recorded, and automatically analyzed digital health marker. There is a need for clinically validated longitudinal, patient-generated (i.e., not health-records based) speech data in psychosis for large-scale sustainable deployment of speech in digital healthcare.

The Diverse International Consortium of Research in Thought/Language & Communication in Psychosis has 110 international members with 26 sites across the world expressing interest in collecting speech samples using a uniform protocol. One of the main aims of this consortium is to create an open source SpeechBank for Psychosis. This project will generate the infrastructure to host this first-ever effort to create an open-access repository for speech data collected during clinical visits. As we develop project-specific funding applications, TOSI鈥檚 support will enable us to (1) create a secure cloud space to hold anonymized audio data from participating sites (2) disseminate a protocol (already developed by our Steering Committee) for harmonized speech sampling across sites that will contribute data to this SpeechBank, and (3) develop a pipeline for uniform pre-processing of audio data for computational analysis by the user community.

Furthermore, this will provide the first crucial step to create a repository for clinical linguistic data in the domain of mental health鈥攁n area where open clinical data is difficult to obtain.

Four projects have been awarded funding through the 2022 Neuro Community Open Science Initiative call supported by the Tanenbaum Open Science Institute (TOSI). Each of these funded projects facilitates the open sharing of scientific outputs with key stakeholders, including other researchers, patients, policymakers, and the public.

• D茅j脿 vu: Development of an openly available knowledge portal for epilepsy presurgical evaluation
  John Thomas (The Neuro), Birgit Frauscher (The Neuro), Jean Gotman (The Neuro)

• The White Matter Rounds: From monthly meetings to Open Science Network
  Roberta La Piana (The Neuro), Sunita Venkateswaran (Children鈥檚 Hospital of Eastern Ontario), Ruwan Bedeir (The Neuro), Stephane Mukunzi (The Neuro)

• Open-source software to automatically detect epileptic spikes on EEG and MEG signals with AI
  Sylvain Baillet (The Neuro), Julien Cohen-Adad (Polytechnique Montreal, CRIUGM, Mila)

• The PynaSuite, open-source software for neural data analysis
  Guillaume Viejo (The Neuro), Adrien Peyrache (The Neuro)

To learn more about each of these Open Science initiatives, please see below.

D茅j脿 vu: Development of an openly available knowledge portal for epilepsy presurgical evaluation

Up to 40% of patients with epilepsy do not respond to anti-seizure medication, and this group requires up to 80% of the resources available for epilepsy care. The only possibility to cure drug-resistant focal epilepsy is surgery. The use stereotactic-EEG (SEEG) for pre-surgical planning has gained increasing popularity but remains limited, and clinicians rarely have access to a large number of patient case studies to optimize SEEG surgery planning and maximize surgical outcomes. This project aims to augment our limited knowledge on SEEGs and create an open, multicenter, Pan-Canadian directory of patients that could optimize pre-surgical planning for focal drug-resistant epilepsy.

The aims of the project are the following: (1) Develop a repository of epilepsy surgery patients based on patient clinical and investigation characteristics; (2) Implement a portal to retrieve information regarding patients with similar characteristics to a new surgical candidate; (3) Disseminate the portal description and details through open access publications; (4) Expand the database by recruiting additional patients through our collaborators from the Pan-Canadian epilepsy network.

The White Matter Rounds: From monthly meetings to Open Science Network

The White Matter Rounds Network is a one-of-a-kind initiative that brings together clinicians and scientists from 16 centers around the world interested in rare and atypical white matter disorders. The WM Rounds were initially local monthly meetings between MS specialists, radiologists, and geneticists to discuss clinical cases of rare white matter disorders that could be amenable for genetic testing, and involve colleagues from 16 centers in North and South America, Europe, and the Middle East. The overarching goal of this project is to formalize and extend the WM rounds in an international network to advance the quality of research in the field of rare white matter disorders by creating an Open Bioregistry of adult genetic white matter disorders that will accelerate the diagnosis of these rare patients. The synergy with patients鈥� associations enhances these collaborative efforts and facilitate access to rare patients鈥� populations.

An open-source software to automatically detect epileptic spikes on EEG and MEG signals with AI

Around 50 million people worldwide suffer from epilepsy, making it one of the most common neurological diseases globally. Epilepsy affects the central nervous system and prevents patients from living a normal life. Electroencephalography (EEG) and magnetoencephalography (MEG) recordings contain patterns of abnormal brain activity such as interictal epileptiform discharges (IEDs), also known as spikes, that are diagnosis signature neurologists are after when analyzing these long recordings. However, such signals are tedious and time consuming to manually analyze. The goal of this project is to create and distribute an open-source robust user-friendly AI-powered pipeline to epileptologists to study brain recordings and localize epileptic spikes.

To achieve the AI-powered analysis pipeline of EEG/MEG signals, we will create interoperability between Ivadomed and Brainstorm software. Brainstorm is a collaborative, open-source software dedicated to the analysis of brain recordings and Ivadomed is an open-source software toolkit allowing researchers to conveniently prototype deep learning models for applications in medical imaging. Physicians and researchers will be able to explore/analyse brain recordings signals (EEG and MEG) on the Brainstorm side, while leveraging powerful AI methods thanks to Ivadomed. This novel open-source software will be tailored to epileptologists and researchers enabling the study EEG/MEG brain recordings and automatic localization epileptic spikes without any programming knowledge.

The PynaSuite, open-source software for neural data analysis

An increasing number of laboratories are using simultaneous calcium imaging, electrophysiology and/or behavioral assays to address their scientific questions. The combination of multiple modalities results in complex data structures, increasing the difficulty of data analysis. Most data analysis toolboxes focus on a single modality. The aim of the project is to pursue the development of PynaSuite, a collection of open-source python software for cross-modal neural data analysis. This project has already been kickstarted with the creation and public release of Pynapple, a light-weight library for neurophysiological data analysis (). The data science ecosystem has widely adopted Python as a standard for programming and we are therefore using the same language.

The aims of this project are as follows: 1) Extend Pynapple and release a complete suite of software tools for the neuroscience community, openly and freely accessible. 2) Develop interactive tutorials and workshop hosted on an open platform and 3) Share knowledge through the organization of workshops to teach data analysis for the largest number.