"The brain is the last and grandest biological frontier, the most complex thing we have yet discovered in our universe."-James D. Watson, co-discoverer of DNA's structure".
Three pound organ, which is self conscious, with around 86 billion neurons forming approximately 100 trillion connections has long been a black box in our scientific understanding. The mere complexity it holds has been beyond our comprehension. Recent innovations in neuroscience have enabled us to better understand and decode the complexities of the brain, potentially leading to a scientific endeavor that could surpass the scale and impact of the Human Genome Project.
To truly appreciate the complexity of the brain, the most advanced LLM has 405B parameters (Llama 3.1). Impressive as this is, it pales in comparison to the human brain's potential 860 trillion connections. What's more, while these LLMs require massive data centers and megawatts of power, the human brain operates on a mere 20 watts - about the same as a dim light bulb. This incredible efficiency, coupled with its processing power, makes the brain a marvel of biological engineering that we've yet to fully comprehend or replicate, and most importantly of all, there is no maxing out for the brain, incredible feats of memory to instances of hyperthymesia (highly superior autobiographical memory) is a proof of this.
We stand at an inflection point in neuroscience, reminiscent of where genomics was two decades ago, where we decided to decode the entire human genome. Just as the cost of sequencing a human genome has plummeted from millions to mere hundreds of dollars, leading to breakthroughs in biology, we are now seeing similar advancements in brain research and this time the innovations are more rapid.
Several factors are converging to make comprehensive brain mapping - connectomics - a tangible goal:
- Advanced Imaging Technologies: New techniques allow us to visualize brain structures at nanometer resolution.
- Exponential Growth in Computing Power: We can now process and analyze vast amounts of brain imaging data.
- Artificial Intelligence and Machine Learning: These tools can identify patterns and connections in brain data that would be impossible for humans to detect manually.
- Interdisciplinary Collaboration: The convergence of neuroscience, computer science, and engineering is creating fertile ground for breakthrough discoveries.
Since its launch in 2013, the BRAIN Initiative has driven the creation of revolutionary tools for brain research. These innovations include CLARITY, a method that makes brain tissue see-through for better visualization; optogenetics, which allows scientists to control specific neurons using light; and Neuropixels probes, ultra-thin devices that can record from hundreds of neurons at once. These advancements are dramatically improving our ability to study and comprehend the intricate functions of the brain, opening doors to significant discoveries in neuroscience.
In 2020, researchers completed the first full connectome of an insect brain - the fruit fly - mapping all 25,000 neurons. Scientists can now grow 3D "mini-brains" in the lab, providing new ways to study brain development and disease. By 2024, we have the most detailed brain map ever created - that of the fruit fly . This groundbreaking work, which mapped 140,000 neurons and over 54.5 million synapses, provides an unprecedented window into how complex brains are wired and function.
A prime example of how technology giants are accelerating neuroscience research is Google's Connectomics team. In 2024, they unveiled the largest and most detailed map of human brain tissue to date, covering just 1 cubic millimetre but requiring 1.4 petabytes of data to encode.
Google developed advanced AI tools to construct an interactive 3D model of the brain tissue, leading to surprising findings such as "axon whorls" and neurons connected by as many as 50 synapses. They've created innovative AI algorithms like flood-filling networks and the SegCLR algorithm to automatically trace neurons and identify cell types.
Building on this success, Google is now tackling the mouse hippocampus, aiming to map 2-3% of the entire mouse brain. This project could generate the largest biological dataset ever collected, estimated at 20,000-30,000 terabytes.
As we look to the future, several key areas of research promise to dramatically expand our understanding of the brain:
- Whole brain connectomics: The next big goal is to map an entire mouse brain at synaptic resolution.
- Brain-computer interfaces (BCIs): Companies like Neuralink and Kernel are pushing the boundaries of direct neural interfaces.
- Neuroplasticity and learning: Understanding how the brain rewires itself could revolutionize education, rehabilitation, and treatment of diseases like depression and PTSD.
- Precision neuromedicine: Brain mapping could lead to individualized treatments for neurological and psychiatric disorders.
- Artificial general intelligence (AGI) and consciousness: Insights from neuroscience are likely to inform the development of more advanced AI systems, like how current knowledge of C. elegans sheds light on forming liquid neural networks.
As our understanding of the brain deepens, it's not just informing neuroscience – it's also revolutionizing artificial intelligence. A prime example of this cross-pollination is the development of Liquid Neural Networks (LNNs), a new AI model that draws direct inspiration from the dynamic, adaptive nature of biological neurons. So, more studies on the brain might help in understanding consciousness, thereby shedding some light on AGI.
The field of BCIs is rapidly evolving, with several key players:
- Neuralink has developed a high-bandwidth BCI with electrodes so thin they must be implanted by a specialized robot. In 2024, they demonstrated their first human subject controlling a computer cursor with thoughts.
- Synchron, backed by Bill Gates and Jeff Bezos, has implanted its Stentrode device in 10 people, delivering the electrode array via blood vessels to avoid open brain surgery.
- Blackrock Neurotech's Utah array has been implanted in dozens since 2004. Their MoveAgain device received FDA Breakthrough Designation in 2021.
- Paradromics is developing a high-bandwidth interface that promises to capture more detailed brain activity.
- Precision Neuroscience, founded by a former Neuralink executive, developed a flexible electrode array that sits on the cortex.
- Motif Neurotech is taking a less invasive approach, with a device that only penetrates the skull to alleviate mood disorders rather than decode complex brain signals.
Each company is balancing factors like invasiveness, data resolution, and specific use cases. While some aim for high-bandwidth communication to potentially decode speech or complex movements, others are focusing on applications like treating depression. Most of these technologies are primarily intended for people with serious physical impairments, but the direction is clear.
As we stand on the brink of this neuroscientific revolution, we are not just observers - we are active participants in one of the most exciting scientific endeavors of our time. The brain, our most precious asset and the source of our thoughts, emotions, and very sense of self, is finally beginning to reveal its secrets.
From curing devastating neurological diseases to expanding the frontiers of human cognition, the future of neuroscience holds the promise of transforming every aspect of human life. We are at the dawn of a new era where the mysteries of the mind are within our grasp. The journey ahead is long and challenging, but the potential rewards - both in terms of human advancement and scientific discovery - are beyond imagination. We're redefining our understanding of what it means to be human.
If you are building in the Brain space, we would love to hear from you and learn from your experiences. Please write to us at shibil.ahammad@specialeinvest.com.
We at Speciale Invest believe in supporting breakthrough technologies that have the potential to solve pressing global problems. As early stage investors, we like to get our hands dirty early on and support founders in their zero to one journey with patient capital, business development opportunities and hiring. We enjoy and thrive on the risk that comes with backing deep-tech startups at the pre-product stage and help through product-market fit, early customers and scale-up. To know more about Speciale’s investments in disruptive technologies, please check our portfolio
Reference
Human Connectome Project. For information on the BRAIN Initiative
The BRAIN Initiative. GooBRAIN Initiativegle's work on brain mapping
Hasani, R. et al. Liquid neural networks. Nature Machine Intelligence 3, 1010–1019 (2021).
Adolphs, R. The unsolved problems of neuroscience. Trends in Cognitive Sciences 19, 173-175 (2015).
