A bleeding edge machine learning platform fashioned through artificial Intelligence is being hailed as a critical new weapon in the fight against coronavirus.

Apollo, developed by Cambridge UK business Agxio, rocketed off the launchpad today.

The Apollo technology operates beyond-human-scale performance, enabling the robotic platform to evaluate critical data to produce predictive models to solve real world problems. It then optimises these to look for patterns or configurations of parameters that human modellers may not even consider or have the patience to develop – all in a matter of hours.

Cambridge Science Park-based Agxio, which also has premises on the Aberystwyth Innovation & Enterprise Campus and is backed by the Welsh Government, is offering free use of the platform, together with its technical support team, to all credible researchers, practitioners and government bodies working to defeat COVID-19 for the duration of the pandemic.

Agxio has specially created a single COVID-19 data portal for the global community. Coviddata.io is open to any parties for augmentation as cases, data and innovations evolve.

Agxio is keen to hear from any data scientists and Python machine learning programmers who would like to volunteer support to researchers’ projects. To put your COVID-19 initiative forward for access to the Apollo platform, or volunteer your technical expertise to projects, contact Covid-19 [at] agxio.com

With the appropriate data, Apollo and the power of machine learning can be used to analyse and predict the efficacy of potential vaccine combinations, outbreak trends, behavioural nudge factors, early warning indicators, medical images against risk indicators and isolation rate projections.

These are just some of the potential applications envisaged by Agxio; it says the full range of use cases for automated machine learning is limitless.

Vitally, the fully automated AI-driven engine doesn’t require the user to be a programming expert or data scientist specialist – enabling an expert in a non-data science or machine learning field to be able to study ideas or data that would otherwise take years of experience to be able to apply.

Agxio CEO and co-founder, Dr Stephen Christie says: “What’s different about Apollo is that this is AI built by AI. It’s the machine building the machines, a series of robots building the best brains to answer targeted questions. 

“Apollo is designed to focus on problems that are beyond human scale in dimension or complexity and is, without doubt, the most advanced approach of its kind.

“What would take a human literally weeks and months to do, Apollo can generate in minutes and hours. Machine learning is one of the most important tools and defining technologies of our generation, and Apollo is a complete game-changer in terms of accelerating the building of machine learning and solutions.

“While humans naturally tend to have biases, Apollo doesn’t have any and is additionally data-agnostic. Most importantly, Apollo is said to have speed and accuracy – and, right now, we need both to be really responsive to the situation. 

“Accurate evaluation of data is vital in the Government’s planning of next-step measures. I think it is critical for the Government to be using the best tools and techniques we have available at this time.

“If you are going to do anything around research and machine learning, data is critical – as is the sharing and pooling of that data in a properly trusted and curated form, and making the data accessible and available to researchers. 

“When making projections on isolation rates and strategies, you need real data and an engine that is able to crunch that data in a structured way, which is Apollo. 

“You  also need the data to be carefully curated and comprehensive. If you don’t have either of those you’re going to struggle to come up with the correct answer.”

Apollo was originally developed as an expert system to enable arable farmers to analyse traditional and advanced IoT data to address the growing population’s needs for improved yields and disease resistance. 

It has since proved to be a powerful tool for a number of different applications including fraud analytics, disease detection, economic anomalies, and bio-sequencing applications – automating the role of the data scientist to build optimal machine learning models against a target prediction. Data-agnostic, it can operate on numerical, textual and image data, both on and off premises.

• To read the latest edition of Business Weekly visit https://bit.ly/2xMd8Di

An exciting young Cambridge company is raising fresh capital and seeking talented new recruits as it bids to revolutionise drug target selection.

Oppilotech Ltd has homed in on what it calls the most critical decision made in drug discovery and developed a high-resolution computational platform based on systems biology and machine learning to model biological processes in cells.

The platform is used to identify first-in-class viable drug targets and develop novel modes of action drugs modulating these. The high level of detail it can generate allows Oppilotech to reveal new biology and identify first-in-class non-intuitive drug targets, It initially focused its modelling efforts on the bacteria E.coli and is in the process of constructing a whole cell model of E. coli based on Machine Learning – a world first.

Oppilotech has several early stage antibacterial drug discovery programmes which it intends to develop towards the clinic and in an exciting development is now also working on modelling pathways involved in cancer.

Co-founder and CEO Ajay Mistry says the company is looking to finalise a seed funding round of £2 million to expand the team, further develop antibacterial programmes and progress working on cancer pathways.

Currently funded by angels such as Sunil Shah and David Ford, Oppilotech has already raised £1.1m. It wants to recruit three computational biologists and a biologist as a matter of priority.

Mistry has a proven track record in building value in biotechnology businesses through assets, platforms and partnerships. Co-founders John George, the chief scientific officer, and Tony Raynham – director of medicinal chemistry – also have stellar CVs. George is a leader in applying systems biology to modelling living cells, building a world-first whole cell model of bacteria based on machine learning.

Raynham has extensive experience in drug discovery gleaned with Millennium, Roche and CRT, focused on oncology. He has taken drug programmes from target discovery into first-in-man trials.

Mistry said: “Drug target selection is the most important decision we make in drug discovery; a selective high-quality molecule will never become a drug if it is modulating the wrong target, which is a key contributor to the high failure rates of drug discovery programmes.

“Current approaches to target selection are based on limited, narrow spectrum analysis of cellular processes. Our vision is to become a leading biopharmaceutical company with a pipeline of first-in-class medicines that will have a meaningful impact on addressing human diseases worldwide. We aim to emulate companies like Galapagos NV.

“When modelling cellular processes, Oppilotech goes into higher levels of details than traditional approaches. Included in our modelling are detailed input parameters like protein abundance levels, mRNA levels, transcription and translation rates, catalytic rates of enzymes etcetera.

“This higher level of detail provides insights to discover new biology and, importantly, viable target selections. We have proved our approach works.”

South Korean Big Data business NHN Corp is investing almost £10 million in  Cambridge UK mobile payments technology specialist Bango to strengthen the companies’ global partnership.

NHN is taking a 4.7 per cent strategic stake in Bango for £3.2m and making an additional £6.5m investment into Bango’s Audiens business; this latter deal will give NHN a 60 per cent stake in the subsidiary.

The vision is for Audiens to become a global data leader through a technologically sophisticated but easy-to-use Customer Data Platform (CDP), which enables brands of all sizes to plan, execute and analyse their online marketing activities. 

This expands an initial relationship between Bango and NHN, announced last October, which enabled app developers and marketers to access anonymised audiences of high value mobile and internet users in South Korea. 

What began as a further licensing deal has evolved into this strategic partnership and investment agreement.

The strategy for NHN and Bango is to continue to develop the Audiens CDP product – which is optimised towards quick data onboarding and self-service by marketers – and incorporate advanced capabilities from the established, deep data technology developed by NHN into the Audiens product.

The £6.5m investment, together with the technology expertise brought by NHN, will enable the existing Audiens business to accelerate its growth as it addresses a larger global market opportunity. 

As well as enriching the Audiens CDP product, this investment gives NHN data technology a new channel to reach global customers beyond South Korea and enables Audiens to bring new products to market more quickly.

Jonathan ‘Jonno’ Brech, previously CCO at Audiens, has been appointed CEO to lead the expanded business along with Marko Maras, founder of Audiens, who will head-up product strategy and innovation. 

Jonno was previously CEO at Cambridge Data, which advised major organisations developing world-class data driven capabilities. He was interim head of digital at KFC UK & Ireland. 

The business will continue to operate from Italy (Milan) and UK (Cambridge), with a new Audiens R & D presence in Seoul. After the investment by NHN, the board of Bango Deep will include Ray Anderson, executive chair at Bango, and JS Lee, Director of NHN.

For the year ended December 31, Audiens’ profit before tax (including inter-company transactions) was €0.45m and it had net assets of €0.64m. 

As part of the transaction, Bango has capitalised into Bango Deep equity approximately £6.5m of inter-company loans from Bango to Bango Deep (which includes the purchase consideration for the acquisition of 100 per cent of Audiens) that were on the Bango Deep balance sheet. Completion of the agreement is expected on or around April 15.

Ray Anderson, executive chair and co-founder at Bango, said: “This is one of the most exciting milestones in the history of Bango. NHN is a hugely successful data, marketing and payments business based in South Korea, which is one of the most technologically advanced countries in the world.
  
“NHN has a fantastic growth story and a history of first-mover innovation which, coupled with the existing data platform know-how at Audiens, will enable us to create a powerful force for technology leadership in online commerce.”

JS Lee added: “Together we will create a global leader in data technology. The NHN strategy is to partner with global leaders to take our research and technology to the world. 

“We are privileged to enter into this strategic partnership with Bango which has built a unique position at the heart of global mobile commerce for the 21st century.”

The partnership with NHN began last year with the inclusion of NHN data in Bango Marketplace and is now accelerating through this investment in Audiens,according to Bango CEO Paul Larbey. “This deeper relationship underlines the Bango vision of becoming the technology behind every payment choice,” he said.

The launch of our Cambridge Commercial Edge research report, the findings of which I discussed last month, in many ways seems like a distant memory, writes Will Mooney, partner at Carter Jonas in Cambridge. 

In just four weeks our world has been turned upside down and I can only hope that you are all safe and sound and keeping well.

I plan to write about the topic that we discussed during our launch – property, the road to net-zero and a carbon-neutral future – partly because once we are through this crisis, attention will again need to turn to these matters, and also because the long-term nature of the challenge provides some sort of focus for us all. 

In 2019, the UK Government set ambitious targets to bring all greenhouse gas emissions to net-zero by 2050. Indeed, an increasing number of countries, states, cities and organisations are committing to carbon neutrality by mid-century, and the Government’s dedication to achieving zero carbon requires an increased pace of change in the supply and use of energy. 

Quite rightly, zero-carbon buildings are now getting the attention they deserve as part of the solution. Presently only one per cent of new buildings are carbon neutral but more than 40 per cent of the UK’s greenhouse gas emissions are generated by real estate. 

Clearly, the built environment has a major role to play in meeting these carbon-neutral objectives. Developers, landlords and occupiers are increasingly focusing their efforts on sustainable initiatives including looking at the energy use of the buildings we occupy and their property estates. 

Because whilst we can design new carbon-neutral buildings, one of the most pressing issues is what to do with the huge amount of existing stock.

My colleague, Greg Hilton of our energy team, discussed just this at our launch event. He is working with clients looking at their existing buildings and portfolios to reduce their carbon intensity and energy demand. They do this by monitoring how much energy a building or estate uses, and how are they employing that energy, to then consider what steps they can take to invest in renewable technology.

As an example, his team has been working with a client with an electricity-intensive building, but an extremely large roof space, which includes an above-ground car park. 

Having undertaken a site screening, feasibility studies and financial modelling he is working with his client to install solar carports and solar panels on the roof. 

Of course, not every client has the capabilities/infrastructure to invest in such initiatives but it is clear that we absolutely have to use less energy and design out heat demand from existing building stock and even new schemes.
 
Other innovations such as the use of LED lights and sensors in buildings enable us all to be smarter. This is software that actively increases efficiency in a building – turning off lights automatically when a room is not in use and adjusting the temperatures as and when required. 

For commercial building owners and occupiers, energy storage is also providing new opportunities. In buildings, battery storage is used to reduce the cost of electricity by maximising the self-consumption of local renewable production or by reducing periods of peak consumption.

Just these few examples demonstrate some of the steps that the property sector can take towards the road to net zero, but there are countless others including encouraging the use of sustainable materials in new-build schemes and designing new buildings that target high BREEAM, LEED and/or EPC ratings, to name but a few. 

Alongside this, new technologies such as mobile apps are being developed that will improve efficiency and the use of space during low times.

The subject matter is vast, but one of the main take-outs from our launch discussion is that the property sector has a critical role in helping to create a sustainable future. 

Our built environment is undoubtedly part of the problem, but if we share best practice and knowledge across the industry, there is a chance that we can form part of the solution and help in reducing that damning 40 per cent figure.

carterjonas.co.uk/

The Information Commissioner’s Office (ICO) recently published its draft ‘Guidance on the AI Auditing Framework’, writes Georgina Perrott, Solicitor with Birketts LLP.

The term AI is used to refer to technologies that mimic human thought, both purely software-based systems like voice assistants, search engines and image analysis software and hardware devices such as autonomous cars and internet of things applications. 

The ICO is concerned about the volume of personal data used by some AI systems and the ability of AI to intrude into private life and effect human behaviour. 

It has previously identified certain characteristics of AI which raise data protection concerns including the use of algorithms for undetermined purposes, the opacity of processing by AI and a tendency to collect all available data. 

Whilst it is not legally-binding, the draft guidance provides what the ICO considers best practice for data protection-compliant AI, with practical advice on how to identify risks posed by AI to individuals and how best to mitigate such risks. 

It is aimed both at organisations that build their own AI systems and those which use AI developed by third parties and focuses on the challenges posed by AI in the following four areas of data protection.

Accountability and governance 
Under the accountability principle organisations are responsible for demonstrating compliance of any AI system with data protection. Interestingly, the ICO recognises that, depending on their design and deployment, AI systems are likely to require making ‘trade-offs’ between privacy and other competing rights and interests. 

For example, using additional personal data in a data set to improve the statistical accuracy of an AI system will enhance compliance with the fairness principle but it will also increase the privacy risk. 

The guidance recommends ways to assess trade-offs. Such assessments should be included in compliance documentation. 

It acknowledges that a ‘zero tolerance’ approach to risks to individuals’ rights is unrealistic and that instead, organisations should ensure that such risks are identified and mitigated. 

Lawful and fair processing 
The guidance highlights the importance of identifying an appropriate lawful basis for each processing operation. This can be challenging in an AI-context. 

It can be difficult, for example, to obtain valid consent for more complicated processing operations where the processing of data may evolve without human intervention. 

The guidance discusses when different conditions, including consent and legitimate interests, may be appropriate for AI systems.

For AI systems which infer data about individuals, statistical inaccuracy and bias are key risk areas in relation to the fairness principle. The guidance clarifies that a system doesn’t need to be 100 per cent accurate but the greater its accuracy the more likely it is to be compliant. 

The ICO stresses the importance of ensuring that data used to train and test AI systems is balanced and the guidance includes techniques to manage imbalanced or biased data. 

As AI systems sometimes become less accurate over time, the ICO recommends that an AI system’s performance be monitored for any inaccuracy or discrimination. 

In addition, to ensure fairness, where an AI system is used to produce a statistically informed guess about an individual (for example that they are a parent) records must indicate that such data is a guess to avoid it being misinterpreted as factual. 

Security and data minimisation 
The complexity of AI systems, reliance on third party code and/or relationships with suppliers, the need to integrate different IP components and the use of large data sets can exacerbate security risks. 

This complexity can also make it harder to identify and manage such security risks. The guidance describes risks that apply particularly to AI, such as model inversion attacks, and security measures to consider implementing.

Under the data minimisation principle only the minimum amount of personal data to fulfil the purpose should be processed. This can appear difficult to comply with for AI systems requiring large amounts of data. 

The guidance explores different approaches to develop functional AI systems that only process the data required and stresses the need for the input of those responsible for risk management at the design and build stages of AI systems. 

Individuals’ rights 
Personal data is often managed in unusual ways in relation to the development and deployment of AI systems. This can make it more challenging to know when and how individual rights apply to such data. 

The ICO provides guidance on this and on the implementation of effective mechanisms to enable individuals to exercise their rights. The guidance also covers fully automated decision making and the safeguards that need to be in place. 

The guidance is still in draft form but it provides a useful indication of the ICO’s current approach to AI systems and what it expects of organisations which are developing and/or using AI systems in order to comply with data protection legislation. 

Ahead of the publication of the final guidance, organisations using AI may wish to use this draft version to review existing practices to check they align with the ICO’s current thinking. 

The guidance recommends that both technology specialists developing AI systems and those with a compliance focus be involved in achieving data protection compliance so when conducting any review, an organisation should seek input from both of these parties.

• You can call Georgina Perrott on 01223 326635 or email her at: georgina-perrott [at] birketts.co.uk

A new rapid diagnostic test for COVID-19, developed by a University of Cambridge spinout company and capable of diagnosing the infection in under 90 minutes, is being deployed at Cambridge hospitals, ahead of being launched in hospitals nationwide.

The SAMBA II machines, developed by Diagnostics for the Real World, provide a simple and accurate system for the diagnosis of COVID-19 infection. 

They will be used by healthcare workers at point-of-care in order to rapidly diagnose patients, directing those who test positive for the infection to dedicated wards. They can also help identify which healthcare workers are infected, enabling those who test negative to return to the front line.

The machines will be made available to a number of hospitals across the country thanks to a $3million donation from the businessman and philanthropist Sir Chris Hohn, which will enable the purchase of 100 machines. 

The donation has enabled Addenbrooke’s Hospital, part of Cambridge University Hospitals NHS Foundation Trust, to obtain the first 10 SAMBA II machines this week for use in wards where suspected COVID-19 patients are brought in. The donation will be matched by the purchase of 10 additional machines by the Cambridge Trust.

SAMBA II looks for tiny traces of genetic material belonging to the virus, amplifies it billions of times chemically and is therefore extremely sensitive in the detection of active infections. 

Dr Helen Lee, CEO of Diagnostics for the Real World said: “Our goal has always been to make cutting-edge technology so simple and robust that the SAMBA machine can be placed literally anywhere and operated by anyone with minimum training.”

Patients will provide a nasal and throat swab. Once these have been loaded into the SAMBA machine, the remainder of the process is fully automated. At the moment, tests are sent for analysis in centralised laboratories and this, compounded by the sheer number of samples that are having to be analysed, means that diagnosis can take one to two days.

 SAMBA II is able to deliver results while the patient waits, helping healthcare workers ensure that those infected can be quickly directed to specialised wards. Whereas current tests can take over 24 hours or longer to deliver their results, SAMBA is able to deliver a diagnosis in less than 90 minutes.

The tests have been validated by Public Health England, Cambridge, in 102 patient samples and shown to have 98.7 per cent sensitivity (ability to correctly identify positive cases) and 100 per cent specificity (the ability to correctly identify negative cases) compared to the currently used NHS/Public Health England test. This has enabled the team to obtain a CE mark.
 
Dr Martin Curran who conducted the evaluation said: “I am extremely happy with the performance of the SAMBA test because it matched the routine centralised laboratory results.”

Professor Ravi Gupta from the Cambridge Institute for Therapeutic Immunology and Infectious Disease, who is leading the ‘COVIDx’ clinical study to evaluate the impact of the test, said: “Testing healthcare workers could help reduce the risk of infection in healthcare facilities themselves, which might in turn assist national control efforts. 

“It will also reduce the number of staff self-isolating for symptoms as we could use the test to determine who is actually infected. At present the lack of testing is resulting in severe staff shortages nationally.” 

Research nurses to support COVIDx will be provided by the NIHR Cambridge Biomedical Research Centre.

Researchers at Cambridge will also be using SAMBA II to test healthcare workers in high-risk areas such as intensive care units or COVID-19 wards. Their aim is to see whether the tests can identify asymptomatic individuals – those who are infected but do not realise it – so that they can self-isolate and prevent inadvertent transmission.

The technology behind SAMBA II was developed while Dr Lee was at Cambridge’s Department of Haematology. The development of the technology has been supported by Wellcome, the Children’s Investment Fund Foundation, the US National Institutes of Health and Cambridge Enterprise, among others.

“We urgently need rapid diagnostic tests to help the NHS and Public Health England manage the coronavirus outbreak and identify those patients at risk to themselves and to others,” says Sir Chris Hohn.

“I’m delighted to have supported Dr Lee’s important research and now help begin the rollout of this cutting-edge technology across the NHS. This is a game changer.”

Science & research colossus Professor Michael Wakelam, director of the Babraham Institute, has died aged 65 from respiratory complications arising from a suspected COVID-19 infection.

Professor Wakelam joined the Babraham Institute as director in 2007. Derek Jones, chief executive of Babraham Research Campus, said everyone at the location was devastated by the loss of a great scientist and researcher.

He said Professor Wakelam brought a dedication to scientific expertise, both in creating and protecting the environment required for excellent science to happen, and in creating an environment that developed expertise and capabilities in each individual to allow them to achieve their best. 

A tribute on the Babraham website added: “He was passionately committed to providing an excellent training experience for the Institute’s PhD students. 

Michael was an active voice on the value of fundamental research and international science. His research passion was lipids (cellular fats) and the techniques used to study them, and he maintained a research lab and lipidomics facility at the Institute during his time as director.

“Professor Wakelam was a strong advocate of the power of uniting academic and commercial research, as exemplified by the fruitful partnerships that exist today between the Institute’s academic research and the commercial research community of the Babraham Research Campus.

“Professor Wakelam’s warm personality and care for others were reflected in his leadership of the Institute. He enjoyed the opportunities his work brought in meeting and connecting with others, including representing the Institute as part of the EU-LIFE alliance of research institutes. His loss will be felt widely across the scientific community and by all those who knew him.”

Professor Wakelam is survived by his wife Jane and their two sons Alex and Patrick.

Stellar career

Professor Wakelam obtained his BSc in Medical Biochemistry (1977) and PhD in Biochemistry from Birmingham University (1980). Following post-docs at the University of Konstanz in Germany and Imperial College London (as a Beit Memorial Fellow) he was appointed to a Lectureship in Biochemistry at Glasgow University in 1985. 

In 1993 he moved back to Birmingham as Professor of Molecular Pharmacology in the Institute for Cancer Studies. He joined the Babraham Institute as Director in 2007.

Professor Wakelam was the Honorary Professor of Lipid Signalling in the Cambridge University Clinical School, an Honorary Professor at the University of Birmingham and a visiting Professor at King’s College London. 

He was a Fellow of the Royal Society of Biology and a member of the Academia Europaea. He received the Morton Lectureship from the Biochemical Society in 2018. 

He had over 20 years research experience in the area of cell signalling and communication; a major focus of his research was upon the use and development of advanced lipidomics methodologies in determining the functions of individual lipid molecular species in the regulation of signalling pathways in normal and cancer cells and in inflammatory responses.

Cambridge Cluster science & technology companies have started the fightback against a virtual UK lockdown with global deals and strategies collectively worth billions.

Arm’s Japanese parent SoftBank has led the way by kickstarting the sale of $41bn worth of assets in a bid to restore former glories. AstraZeneca is investing £80 million in a new therapeutics play from its Cambridge HQ.

Abcam and AVEVA are back on the acquisition trail, with Abcam’s expansion in America spearheading a string of new trade deals between East of England businesses and the US.

A cohort of local companies are part of £multimillion projects to provide vaccines, diagnostic tests, sanitants and a battery of weapons to help the UK conquer the coronavirus, COVID-19. And amid it all comes ingenuity to burn. Redgate Software is using the home working period to launch a new coding club for children.

Also, Business Weekly unveils a 12-strong pantheon of ‘Super Scalers’ – companies in our TTP-sponsored Killer50 list of the hottest science & technology companies in the region who are scaling to fresh heights from already dizzy plateaus.

All this and more – including top quoted companies’ latest results and news of a new breed of spinouts and startups – is in a 42-page special digital edition of Business Weekly available to access FREE on this site today.

It shows just how strong the cluster is proving in adversity – and why our companies are attracting so much attention, even now, from global investors and collaborators.

Doors at physical offices may be closed but windows of opportunity for fresh trade have been left ajar to keep UK and cross-border trade ticking over nicely.

To access the digital edition visit https://bit.ly/2Jlmatz

Cambridge Consultants and Neuro Engineering Technologies Research Institute (NETRI) in Lyon, France have achieved a breakthrough in precision imaging of brain activity. 

Working with NETRI’s brain-on-a-chip technology and lensless imaging approach that detects the activity of neurons in brain tissue, Cambridge Consultants was able to accelerate the processing of individual images by over three orders of magnitude, moving from about 20 minutes to process an image to just hundreds of milliseconds. 

This step-change in speed opens up radical new possibilities, including real-time processing of neural activity and mimicking of neural activity on engineered platforms, supporting the development of novel treatments for conditions such as Parkinson’s and Alzheimer’s diseases.

Organ-on-a-chip is an emerging and highly promising area of biomedical research. The technology involves the ex-vivo (outside of the body) simulation of organ function on a chip. 

The ultimate promise of such technologies is to greatly reduce the time and cost of testing new therapies for safety and efficacy. Brain-on-a-chip technologies seek to apply this approach to the neural circuitry of the brain, the most complex organ in the human body.

French startup NETRI creates disruptive technologies for brain-on-a-chip applications in neuroscience. Its focus is on ex-vivo replication of human neural circuits in order to understand how neurological disorders, treatments and chemicals affect the brain. 

NETRI uses a combination of brain-on-a-chip technology and lensless microscopy to record the activity of millions of connected neurons, mimicking the neural circuits implicated in neurological disorders.

However, by detecting something as complex as neural activity, the approach generates an enormous amount of data, requiring many hours of computation effort in order to fully reconstruct the spatio-temporal map of neuron communication.

The imaging system runs at approximately 1,000 frames per second, and the baseline algorithm took an average of 1,090 seconds – 18 minutes – to process a single image on a powerful desktop PC. Without improvement, a single second of recording can take 12 days to process.

Recognising its world-class expertise in medical imaging and high-performance computation, NETRI approached Cambridge Consultants to address this bottleneck, challenging the company to develop a radical algorithm acceleration. 

A scientific imaging team was set to work within the company’s Data Lab, an on-site facility with state-of-the-art compute infrastructure, allowing rapid exploration of data and compute-heavy approaches, such as high-definition imaging and deep learning.

Cambridge Consultants chose the NVIDIA DGX POD architecture for the AI infrastructure, deploying the NetApp ONTAP AI solution that combines NVIDIA DGX-1™ with NetApp ONTAP storage and network fabric. 

Through a series of optimisations, the team was able to show that the NVIDIA DGX-1 system delivers the AI supercomputing power required to process a frame in an average of 0.3 seconds, around 3,000 times quicker than the original approach. 

This acceleration was achieved through a series of optimisations including the design and application of mathematical and algorithmic improvements, leveraging NumPY GPU-accelerated with RAPIDS (CuPy), and exploiting the sheer power of the DGX-1 system. 

The combined result enables NETRI to completely reimagine their approach and to set their sights on the most radical possibilities, including developing a fully scalable and real-time medical service for hospitals and pharmaceutical industries to perform both in-vitro diagnostics and personalised treatment for every patient suffering from a neurological condition.

Sally Epstein, head of strategic technology at Cambridge Consultants, said: “Lensless imaging replaces traditional optics with computation, creating a new set of optimisation challenges that our team was excited to address. 

“We’re proud to be at the leading-edge of developments in brain-on-a-chip technology and neural imaging, at the point where radical improvements can be unlocked by those with the right combination of vision, experience and compute power.”

Thibault Honegger, chief scientific officer, president &  co-founder of NETRI, added: “By combining the expertise of Cambridge Consultants with NVIDIA-based computational power, the real-time processing of neural communications becomes possible.

“This can mark a paradigm shift in assessing the functional aspects of in-vitro neural networks. With this level of resolution, we will be able to accurately measure the wellness of the network, in relation to neurological conditions and potential recovery.”

Scientists and clinicians from Welcome Sanger Institute in Cambridge and Quadram Institute in Norwich are weighing into a £20 million project to map how COVID-19 spreads and behaves by using whole genome sequencing. 

The Government and the UK’s chief scientific adviser have today backed the consortium which will look for breakthroughs that help the UK respond to this and future pandemics – and save lives.

COVID-19 Genomics UK Consortium – comprising the NHS, Public Health Agencies, Wellcome Sanger Institute, Quadram Institute and numerous academic institutions – will deliver large scale, rapid sequencing of the cause of the disease and share intelligence with hospitals, regional NHS centres and the Government.

Samples from patients with confirmed cases of COVID-19 will be sent to a network of sequencing centres which currently includes Cambridge, Norwich, Belfast, Birmingham, Cardiff, Edinburgh, Exeter, Glasgow, Liverpool, London, Nottingham, Oxford and Sheffield.

Dr Justin O’ Grady will lead Quadram Institute Bioscience’s work on analysing COVID-19 samples and take a targeted sequencing-based approach to the genetic material to help identify any changes in the virus. Head of informatics Dr Andrew Page will lead the bioinformatics analysis at the Quadram Institute at Norwich Research Park.

By looking at the whole virus genome in people who have had confirmed cases of COVID-19, scientists can monitor changes in the virus at a national scale to understand how it is spreading and whether different strains are emerging. This will help clinical care of patients and save lives.

The CLIMB project, which is led by Professor Mark Pallen, research leader at Quadram Institute Bioscience, and principal investigator on the Medical Research Council-funded CLIMB project, will be providing computing support to the national sequencing efforts.

Director of the Quadram Institute, Professor Ian Charles, said: “We welcome this vital work announced by the Chief Scientific Adviser to understand how COVID-19 spreads and behaves by using whole genome sequencing. 

“It’s a testimony to the excellence of the scientific expertise we have here in Norwich that we will be contributing to this national, collaborative effort.
“I am very proud of all the efforts that my colleagues at the Quadram Institute and across the Norwich Research Park are making to reach the scientific answers we need to deal with this pandemic.”

Government  chief scientific adviser, Sir Patrick Vallance said: “Genomic sequencing will help us understand COVID-19 and its spread. It can also help guide treatments in the future and see the impact of interventions.

“The UK is one of the world’s leading destinations for genomics research and development, and I am confident that our best minds, working as part of this consortium, will make vital breakthroughs to help us tackle this disease.”

The UK Consortium, supported by the Government, including the NHS, Public Health England, UK Research and Innovation (UKRI), and Wellcome, will enable clinicians and public health teams to rapidly investigate clusters of cases in hospitals, care homes and the community, to understand how the virus is spread and implement appropriate infection control measures.

Sir Jeremy Farrar, director of Wellcome, said: “By bringing together public health expertise from Public Health England and genomic science from the Wellcome Sanger Institute the UK can crack the code of this virus, and we should give everyone involved huge credit for that.

“Rapid genome sequencing of COVID-19 will give us unparalleled insights into the spread, distribution and scale of the epidemic in the UK. The power of 21st century science to combat this pandemic is something that those going before us could not have dreamt of and it is incumbent on us to do everything we can to first understand, and then limit, the impact of COVID-19.”