Biocomputing

From
Revision as of 09:15, 25 August 2026 by BPeat (talk | contribs) (→‎Organoid Intelligence)
Jump to: navigation, search

YouTube search... ...Google search

Biological computing; also known as wetware computing or Organoid Intelligence(OI), is an emerging technological field that integrates living biological neurons and brain organoids with synthetic hardware, microfluidics, and computational interfaces. By using the natural self-organizing and energy-efficient properties of biological neural networks, researchers build hybrid systems capable of sensory processing, learning, and real-time computation.


Overview

Unlike traditional silicon architectures based on static binary circuits, biocomputing systems use living human or rodent neurons cultured on high-density microelectrode arrays (MEAs). The biological network dynamically rewires its synaptic connections in response to electrical and chemical feedback. This provides massively parallel data processing, high sample efficiency, and adaptive learning capabilities.

Leading Commercial Organizations

Cortical Labs

a Melbourne-based synthetic biological intelligence startup building hardware and software interfaces that turn biological neural networks into programmable, real-time computing systems.

Core Technology and the CL1 Platform

  • Integrated Wetware: Cortical Labs develops the commercial CL1 system, an automated biological computer housing roughly 200,000 living human or rodent neurons cultured directly on high-density microelectrode arrays (MEAs).
  • Life-Support and Incubation: The CL1 features built-in perfusion and environmental controls to regulate temperature and gas exchange, enabling continuous experimentation without standard laboratory incubators.
  • Open Software Stack: The platform provides open-source Python APIs and real-time C/JavaScript bridges that allow software developers to send electrical stimulation patterns to specific electrode coordinates and decode firing spikes into programmatic commands.

Primary Applications

  • High-throughput drug screening and neurological disease modeling (epilepsy, neurodegeneration).
  • Benchmarking biological sample efficiency against deep reinforcement learning algorithms.
  • Low-power biological computational controllers for autonomous robotics.


FinalSpark

a Swiss biocomputing enterprise based near Lake Geneva, co-founded by Dr. Fred Jordan and Dr. Martin Kutter, focused on providing remote, cloud-based access to living three-dimensional human brain organoids.

The Neuroplatform

  • Cloud Wetware-as-a-Service: FinalSpark created the Neuroplatform, allowing global researchers to access and program living neural tissue over the internet.
  • Multi-Organoid Architecture: The platform operates 16 individual human brain organoids grown from induced pluripotent stem cells (iPSCs), each positioned on specialized MEAs to record and stimulate neural activity.
  • Remote Python API: Researchers connect to the Neuroplatform via automated scripts, submitting computational routines and reading real-time electrophysiological data.

Technical Innovations

  • 100-Day Viability: Custom closed-loop microfluidics and life-support incubators sustain organoids in an active firing state 24/7 for up to 100 days.
  • Dopaminergic Photostimulation: The platform uses molecular "cages" containing dopamine. Pulsing UV light at precise coordinates uncages the dopamine locally, providing chemical reward feedback during computational training cycles.

Leading Academic Laboratories

UC San Diego Sanford Stem Cell Institute (Muotri Lab)

The Muotri Lab at the University of California San Diego focuses on human neurodevelopment, evolutionary neurobiology, and embodied biological computing using brain organoids.

Research Focus Areas

  • Closed-Loop Embodiment: Connecting brain organoids to robotic sensors and actuators, testing the ability of living networks to navigate physical mazes and adapt to motor feedback.
  • Astrobiology and Microgravity: Deploying organoid payloads to the International Space Station (ISS) to analyze how cosmic radiation and long-duration spaceflight affect neural development and electrical signaling.
  • Evolutionary & Disease Modeling: Using CRISPR and archaic genomic sequencing to culture Neanderthal-variant organoids, alongside patient-derived organoids for investigating Rett syndrome, Pitt-Hopkins syndrome, and autism spectrum conditions.

Johns Hopkins CAAT (Hartung & Smirnova Labs)

The Center for Alternatives to Animal Testing (CAAT) and associated engineering laboratories at Johns Hopkins University serve as the academic hub for Organoid Intelligence (OI) standards, biomimetic scaling, and pharmacology.

Research Focus Areas

  • Field Standardization & Embedded Ethics: Drafting the international roadmap for Organoid Intelligence and establishing embedded ethical frameworks for donor consent, tissue governance, and computational use.
  • Biomimetic Perfusion Systems: Engineering microfluidic artificial blood vessels and 3D folding polymeric electrode shells (miniaturized "EEG caps") to record from the entire spherical surface of 3D organoids without core cell death.
  • High-Throughput Toxicology & Drug Discovery: Developing standardized, reproducible organoid arrays to screen drug candidates for Alzheimer's disease, dementia, and neurotoxicity, creating reliable alternatives to animal testing.

Comparison of Major Approaches

Organization / Group Lead Researchers Platform / Substrate Primary Focus & Innovations
Cortical Labs Dr. Brett Kagan 2D cultured neurons on CL1 MEA hardware Closed-loop interactive gaming (Pong, Doom); open-source Python API; sample-efficiency benchmarking
FinalSpark Dr. Fred Jordan, Dr. Martin Kutter 16 3D human brain organoids on Neuroplatform 100-day microfluidic life support; remote cloud API; UV-uncaged dopamine reward training
UCSD (Muotri Lab) Dr. Alysson Muotri 3D organoids (archaic, patient-derived) Embodied robotics; ISS microgravity testing; neurodevelopmental & autism disease modeling
Johns Hopkins (CAAT) Dr. Thomas Hartung, Dr. Lena Smirnova Vascularized 3D organoids (~1 cm) OI field definition; artificial vascular perfusion; high-throughput pharmacology

Key Concepts and Underlying Technologies

Biological Neural Networks

A biological neural network (BNN) consists of interconnected populations of living neurons and supporting glial cells (such as astrocytes and oligodendrocytes) that process and transmit information via electrochemical signaling. Unlike static artificial neural networks (ANNs) implemented in silicon, BNNs exhibit continuous structural and synaptic plasticity. They dynamically form, strengthen, prune, or eliminate synaptic connections in response to real-time sensory feedback, enabling rapid learning and complex adaptation with minimal energy expenditure.

Free Energy Principle (FEP)

Formulated by neuroscientist Karl Friston, the Free Energy Principle (FEP) is a mathematical framework stating that any self-organizing system at non-equilibrium steady state must minimize its variational free energy—effectively minimizing informational entropy, uncertainty, or "surprisal." In biocomputing architectures like Cortical Labs' DishBrain, the principle operates as an unsupervised training driver:

  • When the neural network produces an incorrect response (e.g., missing a ball in Pong), the interface delivers chaotic, unpredictable electrical noise.
  • When the network produces a correct response, it receives structured, predictable electrical pulses.
  • Driven to avoid unpredictability, the biological neurons actively rewire their synaptic pathways to produce motor commands that preserve structured feedback and minimize informational chaos.

Microelectrode Arrays

Microelectrode arrays (MEAs) serve as the fundamental bidirectional physical interface between biological cells and digital hardware. An MEA consists of a grid of microscopic electrical contacts that simultaneously record extracellular action potentials (spikes) from nearby neurons and deliver precise electrical stimulation pulses. Advanced biocomputing setups use both planar, high-density CMOS MEAs for 2D monolayers and flexible, three-dimensional self-folding polymer shells ("miniature EEG caps") designed to wrap around 3D spherical organoids to maximize signal-to-noise ratios.

Organoid Intelligence (OI)

the multidisciplinary scientific field established by an international research coalition led by Johns Hopkins University in 2023. OI focuses on scaling 3D human brain organoids grown from induced pluripotent stem cells (iPSCs) into functional computational units. By pairing vascularized, high-density 3D neural tissue with advanced bioengineering, high-density sensor arrays, and machine learning decoders, OI seeks to understand the basic cellular mechanisms of human cognition, model complex neurodevelopmental disorders, and develop energy-efficient biological computers.

Synthetic Biology

Synthetic biology provides the foundational cellular engineering tools required for modern biocomputing. Through techniques like cellular reprogramming (producing iPSCs from adult somatic donor cells), directed differentiation, CRISPR gene editing, and optogenetics, researchers can generate standardized, reproducible batches of specific human neural cell types. Furthermore, synthetic biology allows scientists to insert light-sensitive ion channels or biochemical reporters, enabling non-invasive optical recording and targeted neurotransmitter release (such as UV-uncaged dopamine) to reinforce biological computation.


Key Pioneers and Research Endeavors

Brett Kagan: The Interface Pioneer

Dr. Brett Kagan, Chief Operating Officer at the Melbourne-based startup Cortical Labs, is a foundational figure in biological computing hardware interfaces and neural plasticity demonstrations.

  • The DishBrain Project (2022): Kagan led the landmark project that integrated living neural cultures with a microchip, successfully training the network to play the 1972 Atari game Pong.
  • Validation of the Free Energy Principle: DishBrain provided a real-time experimental demonstration of Karl Friston’s Free Energy Principle in living cells. It proved that self-organizing biological neurons actively alter their synaptic connections to minimize environmental chaos (unpredictable electrical noise) and restore predictable electrical pulses.
  • The CL1 Hardware & Doom: Cortical Labs developed the commercial CL1 system, housing roughly 200,000 living human neurons connected to an open-source Python API. In collaboration with independent developer Sha (Shan Call), the platform expanded from 2D games to the 3D environment of Doom. Sha created software mapping the game's visual feed to sensory electrodes and translated neural motor spikes into movement and firing commands in real time.

Dr. Fred Jordan & Dr. Martin Kutter: Wetware-as-a-Service

Operating near Lake Geneva in Switzerland, Dr. Fred Jordan and Dr. Martin Kutter co-founded FinalSpark, pioneering the commercial transition of wetware from short-term laboratory experiments to continuous remote cloud infrastructure.

  • Life-Support & Microfluidics: Early 3D brain organoids typically survived for only an hour under stimulation. Jordan and Kutter built closed-loop microfluidic incubation systems that sustain and record living 3D human brain organoids 24/7 for up to 100 days.
  • The Neuroplatform: FinalSpark launched the remote Neuroplatform, hosting 16 living human brain organoids connected to the internet. Researchers from universities worldwide log in, run automated Python scripts to stimulate the cells, and read out electrophysiological data.
  • Dopaminergic Training: The platform trains neural tissue using molecular "cages" of dopamine released by targeted ultraviolet (UV) light pulses as a reward mechanism, laying the groundwork for low-power biological cloud computing centers.

Dr. Alysson Muotri: Morphological and Evolutionary Exploration

Based at the UC San Diego Sanford Stem Cell Institute, Brazilian developmental biologist Dr. Alysson Muotri investigates how biological intelligence develops when connected to embodied systems and varied physical environments.

  • Embodied Organoids: Muotri's laboratory links human brain organoids to physical systems, including multi-legged walking robots, enabling the neural cultures to learn locomotion and navigate physical mazes.
  • Space and Extreme Environments: To test resilience and radiation tolerance, Muotri has sent living brain organoid payloads to the International Space Station (ISS) to study the effects of cosmic radiation and microgravity on neural computation.
  • Archaic & Specialized Genetics: Muotri revived archaic genetic sequences from hominin fossil records to grow "neanderthalized" organoids for comparative neural studies. He also develops specialized organoids from donors with autism spectrum conditions (including his son) to map divergence in synaptic connectivity and communication pathways.

Dr. Thomas Hartung & Dr. Lena Smirnova: Organoid Intelligence Standards

Working at Johns Hopkins University, Dr. Thomas Hartung and his wife, Dr. Lena Smirnova, lead the academic standardization and scaling methodologies for biological computing.

  • Organoid Intelligence (OI) Declaration: In 2023, Hartung and Smirnova assembled an international scholarly coalition to publish a foundational declaration establishing the formal scientific field christened Organoid Intelligence (OI).
  • Vascularization and Perfusion Systems: To overcome the diffusion limits of flat 2D neural sheets and spherical organoids, Hartung’s group develops biomimetic artificial blood vessels and perfusion channels. Supplying oxygen directly to interior cells prevents core necrosis and enables the growth of 1-centimeter, 3D organoids (the scale of a mouse brain).
  • High-Throughput Pharmacology: By building 3D virtual navigation tests and obstacle courses for living tissue, the Johns Hopkins team applies trained organoids to automated drug screening for conditions such as Alzheimer’s disease, depression, and epilepsy, replacing animal testing with physiologically relevant human cells.
Retrieved from "https://primo.ai/index.php?title=Biocomputing&oldid=43237"