Latest Posts:

Sorry, no posts matched your criteria.

Follow Us:

Back To Top
The DSV convergence thesis
Scientific discovery forms the core of our modern lives, yet we still live in a world where a 12% success rate on billion dollar pharmaceutical trials is considered acceptable, the time from lab to application is over 17 years and most results can’t be reproduced.
At its core, nature can be explained with maths. Very complex maths that has evolved to incredible levels of efficiency over millions of years, but maths nonetheless. This can be probed and integrated across domains with the right combination of biological, physics and chemistry expertise.
Traditionally, these different disciplines and even sub-disciplines never meet, and worse, often take a rather dim view of each others’ work. Those attitudes are changing rapidly as a result of more collaborative environments; from the unique layout of the Francis Crick Institute and the interdisciplinary work we do at Deep Science Ventures (DSV), to a notable shift in multidisciplinary focused grant funding and much faster open publishing regimes and pull from industry.
This combination of the convergence of the sciences, digitisation and our compounding capability to measure, manipulate, comprehend and predict the rules and building blocks of nature is rapidly turning applied science into engineering. A shift away from the long wait between serendipitous discoveries and eventual application, and a move towards the reconstitution of tools and components across the sciences to achieve an applied solution.
Here we explore the drivers behind convergent science and the ways that these approaches are leading to a proliferation of potentially game-changing solutions.
“‘Scientific convergence’: leveraging the cutting edge capabilities and principles at the boundary of one field to significantly accelerate the advancement of another.”

The tools of convergent science

Firstly, we can now measure and manipulate the world around us in unprecedented detail

Within the last few years we’ve gained the ability to measure in almost unbelievable detail and at relatively low cost. From the coherence and movement of electrons in superconducting materials like those used in computation and storage, to tracking the real time 3D population level communication of bacteria which allows them to act as a single coordinated unit.
This has come from advances in the application of measuring techniques including NMR (monitoring the absorption and re-emission of radiation in magnetic fields to identify molecules), a wide range of methods with fluorescence allowing spatial and temporal monitoring of cellular components, Raman Spectroscopy (light scattering which can reveal molecular level vibrations), gene sequencing (reading DNA), protein sequencing (quantify array of proteins in cell at any given time), micro-electrodes (monitoring activity of neurons and nerves), high-throughput and a host of other techniques.
At the same time, we can now manipulate the world at an unprecedented level and scale, from the automation of lab work, to gene drives to force certain traits into inheritance and control population, to the construction of nano-materials at an atomic level and the dynamic modification of single base pairs using CRISPR/Cas9 in humans.

Secondly, we can now build models that allow us to truly capture, comprehend and predict the dynamics of complex systems

Modelling and simulation aren’t novel, of course. However, they have always suffered from the combination of insufficient measurable data to build a truly representative model, the lack of compute and the challenges of hard-coded algorithms. This has given simulation a bad name in many domains.
This is changing not only because of the abundance of data and direct digitisation from the emerging high-resolution measurement methodologies described above, but because science has been able to borrow many of the core algorithms and compute improvements from other fields. In this world of interchangeable techniques, algorithms used to understand satellite images can largely be repurposed to detect brain tumors on MRI scans, whilst game simulations can be used to predict the evolutionary path of resistance during cancer treatment and optometry has inspired Google’s path towards sustainable nuclear fusion.

In combination this convergent tool-set is driving game-changing shifts towards addressing our biggest challenges

Keeping us healthy for longer: Therapeutics and digital management
Our attempts at modifying disease biology to date have been relatively blunt and consisted mainly of trial and error, both in traditional small molecules, and across emerging protein and live therapeutics as well as the broader management of disease. Emerging approaches are treating biology as a data optimisation problem that can begin to identify and exploit patterns in the underlying maths of nature. When carefully combined with human creativity this is proving to be a very powerful approach.
Companies such as Peptone are combining quantum machine learning, knowledge of the interaction of chemical elements and NMR structural data data to design therapeutic proteins in minutes, a significant step forward from the 30+ years to hand design human insulin. MicroInventa is able to profile bacteria at scale using a combination of microfluidic, multi-omics and machine learning and determine therapeutic properties at a speed that would have taken two years per bacterium just a few years ago.
There has been a proliferation of gene therapy companies, which leveraging digital tools to directly tweak cellular code using nature’s own tools (CRISPR /Cas9). This has been shown to be effective at fixing faulty genes, but targeted delivery remains a challenge and one that needs to combine deep knowledge of the biology, the physics of membranes and the chemistry of mass production in order to solve. Meanwhile the company Conc-R is using evolutionary mathematics, literally drawing the mathematics behind the evolution of natural species from bacteria to elephants, in order to optimise cancer treatment.
Resources for a growing population
As humans we leave a substantial ecological footprint on this planet, from the energy that we use to the food that we eat and waste we produce. In nearly all of these domains we have stuck with old destructive and inefficient ways because we don’t yet have the technology to build less destructive and cheaper alternatives (this is the real world where capitalism is king). But, there is room for hope:
In regards balancing the energy equation we’re close to understanding how to produce low cost solar cells by leveraging the same unbelievably efficient quantum tunneling properties of biological photosynthetic organisms. Meanwhile whilst battery technology is still struggling to keep up with demand, however, we’re beginning to see additive manufacturing and advanced materials applied to the problem with a step change in performance from the traditional focus on the chemistry.
In crop production and optimisation companies such as Indigo Agriculture are defending against crop disease and improving yield not with damaging chemicals but by using high-throughput techniques to identify symbiotic bacteria and applying these strains directly to seeds.
Many groups are now taking on the challenge of producing meat and other animal products without animals, so called “clean meat”. Aside from the moral issues this has the potential to significantly slow global warming. This draws on years of tissue engineering, stem cell biology and even oncology and will require significant cross disciplinary work from materials, fermentation and fluid dynamics to understand how to grow cells in commercial scale bioreactors. Perhaps more interesting still is the move away from animal tissue to genetically adapt other organisms (insects, plants, fungi and algae) to have the same properties as animal tissue, but with significantly lower input costs than animal-based cellular agriculture.
Many of the current challenges around energy, food and the earth’s resources are no doubt currently rooted in human behaviour and political conflict: this fact hasn’t escaped us. However, with vastly better core economics and productivity, these new scientific advances promise to reduce the damage of behaviour rather than requiring that we change our preferences; to create abundance of resources in a way that neutralises political conflict. In this way, there is a meta-moonshot opportunity to align the public and the individual good through scientific innovation. This is core to our theory of change.
A computationally driven world
The hardware of computation and communication by nature require multidisciplinary solutions and draw heavily on inspiration from other areas.
Quantum computation and instant long distance (e.g. interplanetary) communication is an inherently multidisciplinary approach. From the ion traps (radio frequency cages) in development in academia to novel superconducting materials being tested at IBM. Critically, these are unlikely to be the last word, due to the  constraints around needing to be cooled to near absolute zero and high error rates. Nature doesn’t need to be cooled to exhibit quantum effects, and so there a likely a set of materials that are vastly more effective than current approaches.
Several companies are  directly taking direct inspiration from nature, by either building artificial synapses to speed up certain computational operations, deploying real biological neurons to create hyper sensitive sensors to detect bombs and drugs as in the case of Koniku, or connecting directly with the nervous system via novel microelectronic ‘electroceuticals’, a field that Google already has a commercial presence in via Galvani Bioelectronics.
Machine learning, loosely based on how the brain works, is of course prolific and will drive many of the industries of the future including the design of computational hardware itself.
Data storage has now reached the atomic level and spintronics (using the spin of electrons to store data) may even be nearing prime time. Meanwhile many groups are looking at the potential of synthetic DNA as a storage medium with the current record standing at 215 petabytes per gram (that’s all of Google and Facebook’s data in a few drops), however the cost of synthesising and lack of random access (i.e. you have to read the whole thing like tape) means that this is still some way away from application.
Roll-to-toll printed plastic electronics can now act as simple computers for the price of pennies, IBM has produced a 10c mm sized computer and researchers have combined solar and CMOS technology  into one cell creating a mm sized camera. All perfect for edge computing applications such as sensors in health, IoT and agriculture, where we can empower farmers with new knowledge – such as health of their soil – in what has been traditionally a low-tech environment.
Even the ancient art of origami has inspired ultra-cheap plastic wireless communication.
Whilst meta-materials are equipping robotics with the same sensitive (i.e. high resolution) touch that enables animals to manipulate the most delicate of objects, and Radar / LIDAR that doesn’t require any mechanically moving parts, essential for the next generation of autonomous vehicles.

Building at the convergence

At DSV we look for these inflection points, tools and transferable principles and their potential to solve major challenges when mixed in the right way. We ask seemingly bizarre questions like what can we take from the ‘turning back time’ of stem cells and the latest cell imaging platforms to address the challenge of protein misfolding? How can 3D printing and novel materials enable us to better connect the brain and the digital world? How can we take the principles of the oil and chemical industries and put these to work in mass producing clean meat? We apply these across areas from accelerating the speed of therapeutic development to moving away from harmful chemical processes to more effective and cleaner solutions.
This is just the start, in era that may just be characterised as ‘when the sciences converged’, asking unconventional questions at the boundary of intersecting fields gets us towards designing new materials, energy systems and biological solutions from first principles and, with a bit of luck, a long way towards solving many of our most pressing challenges.