Connect to Crick – SME collaboration
Lead Research Organisation:
The Francis Crick Institute
Department Name: UNLISTED
Abstract
Abstracts are not currently available in GtR for all funded research. This is normally because the abstract was not required at the time of proposal submission, but may be because it included sensitive information such as personal details.
Technical Summary
The MRC Proximity to Discovery scheme awards universities funds to help develop new collaborations, and ways of exchanging knowledge and skills. The awards can be used to support activities that promote the value of academic-industry partnership, and enhance academic and industry researchers’ understanding of each other’s needs and capabilities. This may be through people exchanges, creation of technology demonstrators, showcase events, commercialisation workshops and ‘entrepreneurs in residence’ schemes. Such exchanges of knowledge and skills will boost the most fruitful collaborations between UK universities and life science companies.
People |
ORCID iD |
| Title | Sample preparation for integrated microscopies |
| Description | The Electron Microscopy STP (EM STP) at the Crick has excess demand over capacity for large scale correlative imaging. This situation could be solved with enhanced sample preparation techniques leading to automation of imaging workflows. The Crick and Delmic (the world leader in the fabrication and sale of integrated light and electron microscopes) have worked together before on a multi-partner Next Generation Optical Imaging Award that enabled the first SECOM integrated light and scanning electron microscope for biomedical research, built by DELMIC, to be installed at the Crick. The Crick has developed integrated super-resolution light and electron microscopy using the SECOM platform culminating in the first proof-of-principle application of this high resolution correlative imaging approach. Work has continued on the development of miniaturised fluorescence microscopes as smart tools for automatic tracking of regions of interest through a Crick i2i award. However, progress has been hampered by the relatively poor contrast of the samples in the electron microscope. The objective of the project is overcome this poor electron contrast, building on the results of the Next Generation Optical Microscopy Imagining and i2i awards, through exchange of knowledge and skills between the Crick and Delmic. The project has focused on two areas of sample preparation development: • Improving electron contrast in samples for integrated super resolution light and electron microscopy using standard fluorescent proteins • Development of protocols for integrated microscopy using a new osmium-resistant fluorescent probe |
| Type Of Material | Biological samples |
| Year Produced | 2017 |
| Provided To Others? | Yes |
| Impact | 1) Address the final hurdles in optimisation of sample preparation for integrated microscopy (and therefore address a barrier to adoption). 2) Add value to the microscopy market by demonstrating how integrated microscopes can be designed for correlative experiments. 3) Increase the speed and accuracy of correlative experiments for the global bioimaging community. |
| Description | A novel approach to calculating binding affinities with tensorial networks |
| Organisation | Generative Tensor Networks |
| Country | United Kingdom |
| Sector | Private |
| PI Contribution | The central objective of the collaboration between the Crick and GTN is to enable novel technologies to impact real world drug development. In particular, we aim to combine the expertise of Paul Bates' group on molecular dynamics and binding affinity topics with GTN's proprietary quantum mechanics based deep learning technologies, in order to address challenges of drug development with kinase protein targets. The project will also benefit from a data sharing collaboration between GSK and GTN. The central aim of the 7-month collaboration between the Crick and GTN is to enable novel technologies to impact real world drug development. We are therefore not solely looking to investigate a novel computational approach, but are aiming to use it to address specific challenges of drug development with kinase protein targets, in conjunction with the data sharing collaboration between GSK and GTN. The first stage of the collaboration between the Crick and GTN will concern itself primarily with ligand-based predictive networks. In the longer term (mostly beyond the 7 month funding period) we aim to begin incorporating protein family and binding site-specific information, such as linear and three-dimensional sequence motifs, binding cavity size and relative mobility of key residues. Moreover, disease associated mutations, within and proximal to specific kinase binding sites, will be profiled and abstracted to form unique feature space input vectors, thereby opening the way for personalized drug therapy. The reverse process is also of interest to Paul's group, using advanced tensor network models to analyze molecular dynamic trajectories in order to know when to stop running molecular dynamics (often very costly), to achieve an optimal refinement set of coordinates, either in docking studies or protein model refinement. A paper is about to be submitted by the Bates group on this subject using a more classical recurrent deep network approach; however, further enhancement may be possible using GTN's tensor network methodology. Of specific importance to the collaboration is that Paul and his group have extensive experience in collecting and categorising binding affinity data from the scientific literature. With such data, and employing rigorous cross validation, his group uses an array of machine learning methodologies to calculate affinities. In these machine learning protocols, many molecular features are calculated, such as hydrogen bonding potentials and solvent accessibility terms, which are subsequently analysed as to their relative contributions to the affinity prediction accuracy. Recently, the group has been investigating time series data, as for example obtained by molecular dynamics, to further enhance their predictions. Here the main focus has been on designing recurrent deep networks; however, as stated above, tensor networks developed by GTN is likely to be of great value in this context. Specific datasets - In addition to the datasets described in the introduction to be used for ligand based methods, we aim to combine these with target protein crystallography data. There is a very large family of protein kinases, see KLIFS, some of which have been fully resolved in 3D by such experimental methods as X-ray crystallography. Equally, there are many small molecule drugs that are known to bind, but also known to cross react, with members of this family. Paul and his group have investigated the binding and cross binding of such drugs to the kinase family, employing molecular dynamics to predict relative binding affinities, see for example: Can MM-PBSA calculations predict the specificities of protein kinase inhibitors? CS Page, PA Bates Journal of computational chemistry 27 (16), 1990-2007. Once the new methodology is established and exhaustively benchmarked on this extensive protein family and its small-molecule inhibitors, we plan apply it to other protein families in the future - the timeframe for this is probably mostly beyond the 7-month funding period; for example, P53 and its inhibitors. All families investigated will be of medical importance and the aim will not only be to design improved inhibitors but to understand cross blocking of other family members and thereby reducing toxicity levels. Paul and his group make all computer code and datasets developed and maintained available on the laboratory's website: https://bmm.crick.ac.uk. Paul sees the accurate prediction of binding affinities to be an important medical problem and will ensure all protocols developed in this collaboration, once published, are made publically available, thereby facilitating further benchmarking and development by other competing machine learning groups interested in the binding affinity problem. |
| Collaborator Contribution | The central objective of the collaboration between the Crick and GTN is to enable novel technologies to impact real world drug development. In particular, we aim to combine the expertise of Paul Bates' group on molecular dynamics and binding affinity topics with GTN's proprietary quantum mechanics based deep learning technologies, in order to address challenges of drug development with kinase protein targets. The project will also benefit from a data sharing collaboration between GSK and GTN. At a high level, the collaboration will enable novel methods to be developed via a unique combination of the machine learning technology developed by GTN, capable of handling multi modal datasets - and expertise of Paul's group on topics relating to prediction of binding affinities. This will benefit both GTN and the Crick. |
| Impact | N/A |
| Start Year | 2018 |
| Description | Development of cellular activity assays for ALC1 |
| Organisation | Artios Pharma Limited |
| Country | United Kingdom |
| Sector | Private |
| PI Contribution | Artios will benefit from the Crick's expertise in live cell imagine and its knowledge of the DNA Damage Response and telomere biology. |
| Collaborator Contribution | Artios will provide the Crick with a development opportunity to work in a Biotech environment, which will be highly beneficial for the employee involved in defining his future career path. the collaboration will provide access to unpublished tool compounds and will allow the employee to actively participate in the DD process. |
| Impact | The work is to be published. Skill developments for both organisations. |
| Start Year | 2018 |
| Description | Building AI connections and community |
| Form Of Engagement Activity | A talk or presentation |
| Part Of Official Scheme? | No |
| Geographic Reach | Regional |
| Primary Audience | Industry/Business |
| Results and Impact | The Crick is located in King's Cross' Knowledge Quarter - a world class knowledge cluster encompassing organisations from the Digital Catapult to the Design Council, the Farr Institute and the Alan Turing Institute (ATI) to SQW and Springer Nature. The Crick is both a generator of large volumes of biomedical data through e.g. its imaging technologies, and a consumer of large volumes of data e.g. e-health records and bioinformatics data. Analysing such volumes of data can offer mutual benefit to the Crick and other organisations through testing and validation of novel tools and approaches. This has the potential to create new insights into existing data, enhancing the research being undertaken in the Crick, as well as encourage the adoption of these new tools. The objective of funding these activities is to connect Crick researchers with an interest in data (in the broadest sense) with those in other organisations with a similar interest, to stimulate collaboration. Funding and support has been provided for talks and networking events such as: • AI in Biomedicine Special Interest Group (SIG), a joint endeavour between the Alan Turing Institute (ATI), Benevolent AI, the Crick, UCL and Wellcome Trust. The SIG holds quarterly seminars at which there are presentations from both academic researchers and companies, with an opportunity for discussion afterwards. The June 2017 meeting was held at the Crick and featured talks from Dr Markus Ralser (Crick) on the metabolome of a yeast cell predicted from its proteome and Dr Jane Reed (Head of Life Science Strategy Linguamatics) on natural language processing in translational research. The seminar welcomed over 60 attendees evenly split between academia and industry. • the Crick Artificial Intelligence Club (CrAIC) was set up in Spring 2017 with a format of short technical talks each month followed by a networking opportunity. The target audience was initially focussed on post docs in the Crick and its founder HEIs (KCL, Imperial and UCL) but has grown to encompass other local academic institutes and companies. GTN Ltd (with whom we are developing a people exchange) spoke at CrAIC's November meeting. The Crick hosted hosting the London Phd Network's AI symposium on the 12th February 2018 with 240 delegates across several organization academic institute (Crick ; ICR; KCL; QMUL; ATI; UCL) and industry (60 delegates) Separately DeepTech Monday was set up by Dreamstake in Spring 2017 as a monthly networking event aimed at engineers, data scientists and life scientists in academia and industry, venture capital partners, angel investors an deep tech founders at Seed or Series A stage. Cross fertilisation between all these activities has been actively encouraged, and was strengthen with the KQLabs 'challenge-led' event in December 2017 which brought together participants from the SIG, CrAIC and DeepTech Monday. Whilst it is still early days, a growing, vibrant and connected community is developing. There have been initial discussion between Crick researchers and industry partners, including SMEs. The Crick's Electron Microscopy STP recently participated in the ATI's Data Study Group which is now developing into a project involving the ATI. The activities have helped identified which people/groups in the Crick are early adopters and can form the bedrock from which we build our future activity. |
| Year(s) Of Engagement Activity | 2017 |
| Description | KQLabs |
| Form Of Engagement Activity | Participation in an activity, workshop or similar |
| Part Of Official Scheme? | No |
| Geographic Reach | Regional |
| Primary Audience | Industry/Business |
| Results and Impact | KQLabs has held its first co-creation event on 18 December at the Crick. The theme for the event is 'Treating the healthy. How life scientists can work with data scientists to predict and prevent ill health'. The objectives of the event, from a Crick perspective, were twofold: • to facilitate joint working and ideas sharing between Crick researchers and those from other organisations, including SMEs • to seed potential future partnerships Health and wellbeing represents the largest use case for AI and other frontier technologies. This provides humanity with a unique opportunity to shift health service delivery from the sick to the healthy and thus improve the lives of billions. Based around Google's 'Solve for X' approach, the 'moonshot' was a session for experts from diverse backgrounds to visualise what health and wellbeing will look like in 5 years from now. Invitations to participate were circulated through the email lists for DeepTech Monday (over 200 people), CrAIC (~100 people) and the AI in Biomedicine SIG (~300 people) as well as via contacts at Central Saint Martin's, the Royal Colleges of Arts and of Music. Participants (~60) will be selected to ensure a representative mix of career stage, disciplines and organisations - we believe this event will particularly appeal to Crick post docs. The structure of the event was: • experts to share their thinking to paint a picture of the moonshot • participants to form into multi-disciplinary and multi-organisation teams to explore how combining exponential technologies might deliver a 10x improvement in public health. • participants to generate ideas within a framework mapping potential impact with technical complexity. One idea to be selected to be worked up further by the team. The idea to go through guided rapid evaluation and prototyping and a clear plan of action is decided upon. Mentors to be on hand to help get the participants' creativity flowing and guide the process • teams to pitch their ideas back to the whole group setting out the problem they are solving, what the solution is and how they plan to build it, and any technology required/gaps • teams and ideas to be captured for follow on support and development by Dreamstake and the Crick Translation team. Feedback from participants was collected to feed into development of similar co-creation events. |
| Year(s) Of Engagement Activity | 2017 |
| Description | Non-scientific themed networking: BioBeat 2017 |
| Form Of Engagement Activity | A formal working group, expert panel or dialogue |
| Part Of Official Scheme? | No |
| Geographic Reach | Regional |
| Primary Audience | Industry/Business |
| Results and Impact | BioBeat is a collaborative innovation platform for bioentrepreneurs and industry leaders. BioBeat holds its annual event each autumn where business topics for growing the sector are discussed. BioBeat also produces an annual list of movers and shakers in biobusiness. The Crick hosted BioBeat17 on 16th November 2017 with C2C support being leveraged with internal Crick funding and other sources. BioBeat 2017 addressed the theme 'How do we step up biotech partnering to make greater impact on health and wealth?' Almost 200 people from academia, biotechs and pharma, investors, angels and government attended the event. The objective of the event was to explore strategies and opportunities for successful translation between academia and biotech, drawing on personal experience. The event also provided an opportunity for networking between people from different background with similar interests. The event featured a case study of partnership at the Crick presented by Lucy Collinson, the Head of Electron Microscopy STP at the Crick and two panel discussions. Lucy presented her collaborative works with many organisations, from large multi-nationals to start ups, to drive technology development and adoption in electron microscopy. Lucy's talk focussed on overcoming the challenge of combining electron and fluorescence microscopy to better locate cells of interest through collaboration with Kingsview Optical and RMC Boeckeler. The first panel discussion, chaired by Veronique Birault, addressed the topic of 'Catalysing biotech partnering opportunities with research institutes' and include contributions from: • Lucinda Crabtree, Senior Investment Analyst, Woodford Investment Management • Jane Dancer, CBO, F-Star Biotechnology • Sarah Howell, CEO, Arecor • Natalie Mount, CSO, Gammadelta The panel explored the value of collaborating with academia, the barriers to engagement for SMEs and biotechs and discussed potential solutions. The second panel discussion, chaired by Kate Bingham Managing Partner, SV Life Sciences focussed on 'Partnering and careers permeability' and feature • Barbara Domayne-Hayman, CBO, Autifony, Chair, Puridify • Julie Simmonds, Director, Equity Research, Panmure Gordon • Andrea Spezzi, CMO and Co-Founder, Orchard Therapeutics • Angela Russell, Associate Professor Medicinal Chemistry, University of Oxford and Co-founder, OxStem Veronique Birault is writing an article Feedback from Crick researchers has been very positive - they particularly welcomed the opportunity to see the broader translation and collaborations landscape and hear from such experienced leading entrepeneurs. The event contributes to building a culture of industry collaboration at the Crick. |
| Year(s) Of Engagement Activity | 2017 |
| Description | Technical Workshop with Horizon Discovery |
| Form Of Engagement Activity | Participation in an activity, workshop or similar |
| Part Of Official Scheme? | No |
| Geographic Reach | Regional |
| Primary Audience | Professional Practitioners |
| Results and Impact | Horizon Discovery is a cutting-edge gene editing company with deep expertise in the CRISPR area. It offers applications that are relevant to the Crick such as gene editing (constitutive and conditional knockouts, knock-ins (point mutations and tags)) in cells and mice, arrayed / pooled CRISPR knockdown phenotypic screens and CRISPRa (activation) and CRISPRi (inhibition) phenotype screens. The Crick has an active Genome Editing Forum (with over 140 members), which brings together researchers to discuss ideas, breakthroughs, protocols, answer questions, share frustrations, and promote collaboration. Funding was provided to support a focussed technical workshop (held on 21 August). The objective was to share Horizon Discovery's expertise in areas relevant to Crick research and STPs and provide Crick researchers with an opportunity to meet the Horizon Discovery team, with a view to generating interest in collaboration. Four members of the Horizon Discovery team, including the CSO and Head of Translational Science, and the Team Leader for Discovery Screening attended the workshop. Over 40 Crick researchers attended the workshop, with a mixture of Group Leaders, post docs, PhD students, STP Heads and their teams. They heard about functional genomic screening, haploid human cells and synthetic lethality. There were a number of in depth discussions about particular aspects of screening and gene editing in mice. The workshop met its objective. As a result of the workshop we are following up a number of leads: • exploring a people exchange between the Crick's Genetic Manipulation Service and Horizon Discovery in the area of gene editing in mice • looking at an exchange of people, and also a potential part-time secondment to the Crick, in screening as the Crick looks to develop its in-house capacity and capability • further workshop and trouble shooting session held with Horizon Discovery |
| Year(s) Of Engagement Activity | 2017 |