Inside the
Nemesis project
The NEMESIS project is a collaborative effort to reframe our understanding of the injured brain and transform how it is treated. This research initiative brings together several fields: neuroscience, clinical neurology, computational modeling, and bioengineering to tackle a major puzzle: why do some parts of the brain fail to recover after stroke, even when they are distant from the lesion and appear structurally intact? The working hypothesis of the Nemesis project is both concise and elegant: neurons surrounding the stroke lesion enter a sleep-like state, disrupting communication between distant brain areas and thereby preventing behavioral recovery of the patient.
Over six years, with the collaboration of four research groups in Spain and Italy, the NEMESIS project seeks to understand this dormant state of neurons after brain injury across multiple levels of analysis from slow-wave activity in laboratory brain slices, altered brain connectivity in mice and human participants with brain lesions to personalized virtual brain models built from real human neuroimaging data. Beyond understanding the mechanisms underlying these “dormant” circuits, the NEMESIS project aims to design treatment protocols to reawaken these circuits, paving the way for new forms of post-stroke therapy that are non-invasive, targeted, and grounded in biologically meaningful brain network dynamics. The NEMESIS project is built on four independent research areas, or “IV pillars,” each crucial for achieving its ambitious goals.
Pilar I: Effects of stroke lesions on brain networks, behavior and recovery
Traditionally, scientists thought that the effects of strokes and other brain injuries were due to direct damage at the lesion site. Because strokes can occur in many different parts of the brain, the resulting symptoms seemed diverse. However, research led by Maurizio Corbetta has shown that this variety is only apparent: hundreds of strokes in different locations actually produce a small set of common behavioral deficits that follow predictable patterns of recovery. In addition, these behavioural patterns are accompanied by a limited set of structural and functional abnormalities in the brain, measurable with advanced magnetic resonance imaging techniques.
Pilar II: Brain injury and sleep-like slow-waves
In an independent area of research, using electrophysiological recordings and stimulation (TMS-EEG), work led by Marcello Massimini showed that brain activity in awake, healthy people displays a rich and complex pattern of activity. However, this pattern collapses into simple slow-waves in states such as sleep, anaesthesia, and coma. Importantly, these sleep-like slow-waves do not only emerge in states of reduced consciousness, but also after brain injuries such as severe traumatic brain injury and stroke. This discovery led to the hypothesis that the brain disconnection and behavioural patterns identified in Pillar I may be caused by underlying slow-wave activity.
Pilar III: Cellular origin of sleep-like slow-waves
Even though slow-wave activity is detected in stroke patients and may be responsible for the behavioural deficits, human data does not allow investigation of the mechanisms generating these slow-waves. Work led by Mavi Sanchez-Vives described the cellular mechanisms underlying slow-waves using several in-vitro approaches, such a local cortical circuits. The same slow-waves observed in unconscious brains and brains after stroke can also be seen in isolated cortical slices in the laboratory. Importantly, these investigations showed that slow-wave activity in the laboratory can be enhanced or suppressed using various stimulation methods, including administration of neurotransmitters, anesthetics, and electric fields.
Pilar IV: Whole-brain computational modelling
Understanding how slow-wave activity can be altered in laboratory conditions still leaves the question of where and how such stimulation should be applied in the human brain. Essential to identifying best targets in the human brain is the ability to simulate the effects of treatment by computational means. Gustavo Deco is a pioneer in whole-brain computational modelling, a framework that simulates brain signals and dynamics using biologically realistic maps of anatomical brain connectivity. This approach successfully simulated and explained brain patterns observed with various neuroscientific techniques (e.g., neuroimaging, EEG) and can greatly aid in identifying optimal stimulation targets in the patient brain.
Integration of the four pillars in the Nemesis project
By integrating these four pillars of expertise, NEMESIS sets itself apart by providing a mechanistic understanding of stroke-related changes across multiple spatial and temporal scales, enabling the creation of individualised computational models of brain dynamics in lesioned brains. The integration of neuroimaging, sleep-wave, and behavioural data from human subjects and rodents will be carried out through Pillars I and II. Pillar III, understanding the cellular and circuit mechanisms, is critical for explaining how slow waves are generated, how they propagate, and, most importantly, how they can be modulated through interventions. Data from pillars I–III will then inform whole-brain computational models in Pillar IV, allowing predictions of the effects of individual lesions in humans and guiding optimal stimulation strategies. Finally, these predictions will feed back into proof-of-concept stimulation treatments for stroke patients, evaluated under Pillars I and II.
This iterative, synergistic approach will provide unprecedented insights into the mechanisms of stroke injury and support the development of treatments that may be applicable not only to stroke but also to other neurological disorders.
Timeline
2024-2025
(ongoing) – Experimental studies begin
Launch of experimental activities across partner institutions (in vitro, in vivo, and modelling).
2025-2026
(ongoing) – Experimental studies continue
Launch of experimental activities across partner institutions (in vitro, in vivo, and modelling).
2027-2028
(projected) – Modelling and neuromodulation development
Development of personalised brain models and neuromodulation strategies.
2028-2029
(projected) – Clinical trials
Potential clinical trials and validation of stimulation protocols.
Project goals
Goal 1: Mapping stroke-induced disconnection and sleep-like brain dynamics
Understanding how brain injury–related disruptions are linked to sleep-like slow wave states. This research will include studies in first-time stroke patients, complemented by animal models and in vitro approaches.
Goal 2: Perturbation of disconnected regions and slow-wave states
Investigating how slow-wave dynamics influence neuronal communication through causal perturbations in humans, animal models, and cortical slices.
This includes developing protocols to modulate slow-wave activity in a controlled and targeted manner.
Goal 3: Develop personalised digital brain models and Reactivate dormant neural circuits through precision brain stimulation
Developing a generative model to predict patterns of disconnection and slow-wave propagation resulting from stroke lesions. Creating patient-specific computational brain maps to guide optimal stimulation and the development of novel interventions.
Our research centres
University of Padua (Università degli Studi di Padova), Italy
The research team led by Prof. Maurizio Corbetta at the Padova Neuroscience Center and the Veneto Institute of Molecular Medicine investigates how stroke affects the human connectome. Stroke induces focal brain damage that leads to widespread network alterations across the brain. A central hypothesis of the group is that these remote abnormalities are linked to the slowing of brain rhythms observed after stroke. The Padova team aims to demonstrate this relationship for the first time and subsequently explore ways to modulate these rhythms to reduce stroke-related disability. This work is complemented by parallel studies in animal models, where similar mechanisms are examined using more invasive recording techniques.
University of Milan (Università degli Studi di Milano), Italy
The research group Integrated Thalamo-Cortical functions research group, led by Prof. Marcello Massimini at the department of Biomedical and Clinical Sciences of the University of Milan (Italy), explores the effects of post-lesional sleep-like slow waves in disrupting local and large-scale functional networks and behaviour. They combine experimental work in neurological patients affected by focal and multifocal brain injury of various aetiologies (vascular, traumatic, and post-surgical), as well as modelling work.
Institut d’Investigacions Biomèdiques August Pi i Sunyer (IDIBAPS), Hospital Clínic, Barcelona, Spain
The research team led by Prof. Mavi Sanchez-Vives investigates how focal brain lesions impact local and global neural dynamics. Their central hypothesis is that affected areas enter a sleep-like, dormant state that prevents effective communication, contributing to cognitive and functional deficits. By reactivating these circuits, they aim to support recovery.
Universitat Pompeu Fabra, Barcelona, Spain
The research team led by Prof. Gustavo Deco focuses on developing large-scale mechanistic models of brain function to understand connectivity disruptions after injury. Within the NEMESIS project, they use whole-brain computational modelling to investigate the mechanisms by which structural disconnections, such as those caused by stroke, disrupt large-scale brain dynamics. By uncovering how these alterations emerge from the interaction of brain structure and local dynamics, they aim to identify principles that can predict mitigation and recovery trajectories and guide targeted interventions to re-establish healthy communication across brain networks.
Stay up to date
Follow the project’s progress