Meningsfulla steg mot Organoidintelligens tas

Artificiell intelligens (AI) har blivit ett välbekant begrepp i den teknologiska världen idag och formar många aspekter av våra dagliga liv. Det finns dock en ny front som växer fram inom vetenskapen – Organoid Intelligence (OI) – som, även om den inte är lika välkänd som AI, har en anmärkningsvärd potential. Men vad exakt är Organoid Intelligence, och varför får den så mycket uppmärksamhet i vetenskapliga kretsar?

The Institute of Electrical and Electronics Engineers (IEEE) definierar Organoid Intelligence som ett ”kollektivt arbete för att främja användningen av hjärnorganoider – små sfäriska massor av hjärnvävnad odlad från stamceller – för beräkning, läkemedelsforskning och som en modell för att på liten skala studera hur en komplett hjärna kan fungera.”

Organisationen pekar också på två banbrytande forskningsinsatser som gjort utvecklingen av organoider möjlig. Vi kommer att återkomma till dessa studier strax. Men innan dess ska vi fördjupa oss i vad den vetenskapliga gemenskapen tror om Organoid Intelligence:s potential.

Ett engagerat forskarlag tror att OI‑baserade bioberäkningssystem kommer att möjliggöra snabbare beslutsfattande. De kommer att kunna lära sig kontinuerligt under uppgifter och införa större energi‑ och datat effektivitet i processen. Dessutom skulle de hjälpa till att förstå patofysiologin bakom allvarligt skadliga utvecklings‑ och degenerativa sjukdomar såsom demens. Självklart skulle de också underlätta behandlingen av dessa sjukdomar genom att identifiera nya terapeutiska tillvägagångssätt.

Med tanke på OI:s enorma förmåga att göra gott, tas flera meningsfulla initiativ mot Organoid Intelligence (OI). Vi kommer att titta på dem i de kommande segmenten. Men vi börjar med de två banbrytande forskningsinsatserna om organoidutveckling som vi redan nämnt.

Induced Pluripotent Stem Cell‑Derived Organoids

IPSC‑deriverade organoider kom in i diskussionerna under COVID‑19‑utbrottet. En vetenskaplig forskningsartikel listade detaljer om IPSC‑deriverade luftvägsorganoider, hjärnorganoider, tarmorganoider, leverorganoider, blodkärlsorganoider, njurorganoider, inneröronorganoider med mera. Med denna variation kommer vår diskussion nu att fokusera specifikt på hjärnorganoider.

Artikeln listar tre avgörande punkter: 

• För första gången utvecklade forskare hjärnorganoider med spontan differentiering av embryoidkroppar (EBs).
• De första hjärnorganoiderna var mycket heterogena.
• Heterogeniteten ledde till att forskarna utvecklade regionsspecifika hjärnorganoider genom att påverka specifika signalvägar.

En annan forskning som bidrog till organoidutveckling rörde 3D‑cellodlingstekniker.

3D Cell Culture

Motivet bakom utvecklingen av 3D‑cellodlingstekniker kom från de brister som observerades i odling av celler i två dimensioner. Tvådimensionella cellodlingsmetoder misslyckas med att reproducera vävnadens anatomi eller fysiologi för informativ eller användbar studie.

3D‑cellodlingstekniker – å andra sidan – ligger närmare djur‑ och mänsklig fysiologi. Och när de blir mer mogna kommer de att hjälpa till att designa och utveckla ko‑odlingar. Dessa tekniker kan visa sig användbara vid ingenjörs‑tillverkning av vävnader för klinisk leverans och vid utveckling av modeller för läkemedelsscreening.

Sådana framsteg inom 3D‑cellodling är också avgörande för forskningen kring IPSC‑deriverade organoider och organoidintelligens. IPSC‑celler kan till exempel utvecklas till vilken cell som helst i ett djurs kropp.

3D‑odlingstekniker och 3D‑scaffolding‑metoder hjälper specifikt biologer att odla IPSC‑deriverade nervvävnader vertikalt och horisontellt, vilket banar väg för utvecklingen av de inter‑neurala nätverk som observeras i ett djurs hjärna.

Att reproducera detta intrikata nätverk möjliggör djupare insikter i organoidintelligens och utveckling av mer sofistikerade metoder.

Brain Organoid Reservoir Computing for AI

Research published in the journal Nature Electronics points out the achievements registered in brain-inspired computing hardware that emulates the structure and working principles of the brain. Researchers believe that this development has the potential to address the current limitations of AI. 

The research, in particular, reports an AI hardware approach that leverages adaptive reservoir computation of biological neural networks in a brain organoid. The approach goes by the name Brainoware. It pivots on performing computation by sending and receiving information from the brain organoid through a high-density multielectrode array. 

Organic Semiconductor Synthesis: Opening Up New Horizons for Super-Efficient Electronic Devices

A team of researchers at UNIST’s Department of Chemistry facilitated the space to further open up to innovation opportunities achieving a breakthrough in organic semiconductor synthesis. The researchers reported the synthesis and characterization of a B2N2 anthracene derivative with a BNBN unit formed by converting the BOBN unit at the zigzag edge. The findings are expected to help develop larger acenes with multiple BN units that will have many applications in organic electronics. 

According to Songhua Jeong, the first author of this study:

“Our study on anthracene, a type of acene widely recognized as an organic semiconductor, has laid the groundwork for future advancements in the field.” 

Mr Jeong further went on to say: 

“The continuous BN bonding synthesized through this research holds great potential for applications in organic semiconductors.”

With diverse research opening up avenues for Organoid Intelligence to prosper, experts in this field say that organoid intelligence may become powerful enough to power our computers in the future. 

A future does not appear unrealistic anymore, where lab-grown brain cells would store and retrieve large amounts of data with smaller demands for energy. With further scale-up achieved, these solutions may even develop a consciousness to think or feel of their own. 

Organoid Intelligence: Workshops and Funding

Diverse efforts are underway to deliberate further on the potential of organoid intelligence. 

Between February 22nd and 24th, 2022, John Hopkins University held the first Organoid Intelligence workshop. The purpose was to form an OI community and formulate the foundation for establishing OI as a new scientific discipline. The motivation behind the workshop came from the belief that OI could radically alter computing, neurological research, and drug development. 

Further validation of the potential of OI came when the United States National Science Foundation Directorate for Engineering planned one new area for the Emerging Frontiers in Research and Innovation (EFRI) program in FY 2024, and it was called ‘Engineering Organoid Intelligence.’ 

The NSF is known for supporting science and engineering research and people related to it by providing facilities, instruments, and funding. The NSF had a budget of US$9.5 billion in the fiscal year 2023. These funds reach all fifty states in the US through grants given out to nearly 2,000 colleges, universities, and institutions. 

The ‘Engineering Organoid Intelligence’ topic is aimed at supporting fundamental research on the development of engineered biocomputers through brain organoids. The NSF believed that Organoid cultures would soon have interfaces with electronics to deliver input and export output to external digital systems.  

Apart from universities, research institutions, and government departments, many companies have taken up the cause of advancing organoid intelligence. We will now look into some such companies and their endeavors. 

#1. Final Spark

Founded by Martin Kutter and Fred Jordan in 2014 and based out of Switzerland, Final Spark advocates for biological chip processors that consume much less energy. 

As claimed by the startup, it has tested 10 million neurons already in its endeavor to build thinking machines from live human neurons derived from skin. 

The startup is leveraging sophisticated cell-culturing techniques to exhibit the self-sustaining computing capability for the creation of future AI models. 

#2. BiologiC and NVIDIA

BiologiC Technologies, a company based out of Cambridge, UK, is known for its biocomputers, which the company defines as an “automated programmable and reconfigurable bioprocessing platform for multiple biological applications.” 

The solution’s patented hardware architecture has four core modules: 

• The advanced routing core helps transport liquids and biogases through the system. 
• The input and output block helps control the loading of raw materials and the output of the final bioproduct. 
• The BPU model helps with bioprocessing activities such as cell growth, in-vitro transcription, etc. 
• Finally, the sensor core module helps with real-time monitoring of the bioprocess environment. 

On 13 mars 2023, BiologiC Technologies joined NVIDIA (NVDA ) Inception, a program that will help BiologiC integrate edge computing and AI into its biocomputers. This integration will eventually help BiologiC’s biocomputers bring revolutionary improvements in its biology processing facilities and reduce the threat of diseases in a sustainable, accessible, and affordable way. 

According to available information, CPI Enterprises invested in BiologiC. CPI invests in high-potential deep tech for companies to scale up fast, playing the dual role of an investor and a technical support provider. 

According to Frank Millar, the chief executive of CPI: 

“Having collaborated closely over an extended period, we are impressed at the speed of execution and commercial traction of BiologIC. We see strong advantages in BiologIC’s bio-computer system and the ability to increase bioprocess productivity relative to the competition in the sector.”

On 21 november 2023, NVIDIA reported revenue for the third quarter ended 29 oktober 2023, of $18.12 billion, up 206% from a year ago and up 34% from the previous quarter.

#3. BioMap and MBUZAI: Joint Efforts on Creating a Biocomputing Lab 

On 14 mars 2023, BioMap struck a strategic collaboration with Mohamed bin Zayed University of Artificial Intelligence (MBZUAI) to introduce the first biocomputing innovation research laboratory in the Middle East and address the needs of life science scenarios in the region. 

The cooperation would explore BioMap’s cross-modal bio-computing model, xTrimo (Cross-model Transformer Representation of Interactome and Multi-Omics), in achieving large-scale life science models in protein generation, protein structure prediction, cell function prediction, and other life science tasks. 

The focus will be on achieving breakthroughs in AI-generated proteins (AIGP) to help with the Middle East’s need for medical health, drug design, energy, and environmental protection.

Apart from protein design, BioMap’s capabilities will also help achieve strong cell characterization capabilities, precise first-time modeling of immune cells, and more. 

Co-founded by Baidu CEO Robin Li, BioMap helps combine best-in-class AI and biotech capabilities to combat the most crucial problems in the life sciences industry, such as target discovery, de-novo drug design, and enzyme optimization.

The biological computing Chinese startup raised more than US$100 million in series A funding, according to reports published in augusti 2021. The round was led by CGV Capital. Other investors included Baidu, Legend Capital, Bluerun Ventures, and Xianghe Capital.

Which Way OI is Headed to?

Undoubtedly, OI has captured the scientific community’s imagination. Sophisticated technological companies and minds are investing resources as well. How the future will shape out for OI would, however, depend on a few crucial aspects that relate to its developmental nature. 

The success of OI as a revolutionary technological innovation would depend on how fast, efficient, and powerful it might become in comparison with silicon-based computing and AI. Moreover, enhanced standards of speed, efficiency, and power would have to have lower demands for energy to make it more attractive and cost-effective. 

In the future, OI needs to become truly multidisciplinary since it is a highly complex and intricate exercise to scale current brain organoids into durable 3D structures. These structures will be full of cells and genes that help humans learn new, complex things. 

Moreover, devising the ideal biocomputers would require connecting these organoids with cutting-edge input and output devices, particularly electronic ones equipped with high-end AI and machine learning systems. 

Researchers in the field believe that there is a great need for new and sophisticated models, algorithms, and interface technologies to establish effective and meaningful communication with these brain organoids. To make the output best-in-class, creators need to know how they learn, compute, process, and store the data. 

The scope of data, as evident already, will be massive and immense in volume. 

In the interim period, OI research will also help improve our understanding of how the brain and its deep and intricate neurological network functions. These understandings will help treat patients who suffer from neurological diseases better.

Finally, replicating the capabilities of a human brain in a lab environment could become an achievement with humongous consequences. The potential for misuse could be significantly high. Therefore, from the very early days, the scientific community must ensure that research and development around OI stay very sensitive to our society’s ethical systems. The ethical framework should be no less than a defining paradigm for OI to thrive in the future. 

Click here for the list of the most promising biotech stocks.