The genetics underlying imaging phenotypes and correlated physiological measures
Lead Research Organisation:
University of Edinburgh
Department Name: Sch of Molecular. Genetics & Pop Health
Abstract
Imaging techniques are widely used for early detection and diagnosis of disease. Although widely adopted in clinical practice, it is only partially understood how structural information in medical images relates to physiological function, which may amenable for treatment. Furthermore, our understanding of how imaging phenotypes are determined by genetics and to what degree the genetic control of disease and the imaging phenotypes is shared is very limited. Taken this into account, this project is focused in understanding how genetics control imaging phenotypes and correlated physiological measures. Specifically, structural information and functional information will be obtained from more than 500,000 fundus images.
Identifying which imaging phenotypes have genetic correlations (i.e. a genetic shared component) with the disease of interest would inform of shared control pathways and to match disease-subtypes and imaging phenotypes to different biological pathways. Furthermore, computational modelling approaches can be used to convert structural images into physiologically relevant traits. During this project we will develop both structural measurements and modelling-based estimates of blood flow (i.e. haemodynamics) in the eye from retinal images in UK Biobank, a large prospective epidemiological study of 500,000 individuals. Another objective of this project is knowing whether these low-level phenotypes have a bigger genetic contribution than high-level phenotypes. Our goal is to identify genes influencing the structural and haemodynamic measurements and dissect their genetic and environmental variation. Further studies will allow us to analyze if these genes could have a medical application, for instance, a druggable gene or a genetic biomarker.
In addition, one of the challenges of this project is the high number of samples in the project because each sample must be analyzed and then merged with all the other results. So, the purpose of this study, in addition to the knowledge obtained from the fundus images and its genetic analysis, is develop a new approach that combines different types of information towards an objective, such as early diagnosis, periodical revisions and stablishing new therapeutic targets.
Identifying which imaging phenotypes have genetic correlations (i.e. a genetic shared component) with the disease of interest would inform of shared control pathways and to match disease-subtypes and imaging phenotypes to different biological pathways. Furthermore, computational modelling approaches can be used to convert structural images into physiologically relevant traits. During this project we will develop both structural measurements and modelling-based estimates of blood flow (i.e. haemodynamics) in the eye from retinal images in UK Biobank, a large prospective epidemiological study of 500,000 individuals. Another objective of this project is knowing whether these low-level phenotypes have a bigger genetic contribution than high-level phenotypes. Our goal is to identify genes influencing the structural and haemodynamic measurements and dissect their genetic and environmental variation. Further studies will allow us to analyze if these genes could have a medical application, for instance, a druggable gene or a genetic biomarker.
In addition, one of the challenges of this project is the high number of samples in the project because each sample must be analyzed and then merged with all the other results. So, the purpose of this study, in addition to the knowledge obtained from the fundus images and its genetic analysis, is develop a new approach that combines different types of information towards an objective, such as early diagnosis, periodical revisions and stablishing new therapeutic targets.
People |
ORCID iD |
| Albert Tenesa (Primary Supervisor) | |
| Ana Villaplana Velsaco (Student) |
Publications
Villaplana-Velasco A
(2021)
Decreased retinal vascular complexity is an early biomarker of MI supported by a shared genetic control
Studentship Projects
| Project Reference | Relationship | Related To | Start | End | Student Name |
|---|---|---|---|---|---|
| MR/N013166/1 | 30/09/2016 | 29/09/2025 | |||
| 2105835 | Studentship | MR/N013166/1 | 31/08/2018 | 30/11/2022 | Ana Villaplana Velsaco |
| Title | Automatised management of large fundus images database |
| Description | We automatised and increased the speed of the management, segmentation and analysis of large imaging databases. |
| Type Of Material | Data handling & control |
| Year Produced | 2019 |
| Provided To Others? | No |
| Impact | This algorith accelarates the quality process, data management and data obtention of medium and large imaging datasets. In our case, fundus image's classification, segmentation and analysis took less than half of the time than traditional process. |
| Title | Myocardial infarction risk stratification in UK Biobank |
| Description | A predictive model for myocardial infarction was developed using 526 UKB participants that had an incident event and had a good quality fundus image and an equal number of participants without MI and associated risk factors. This model included clinical variables traditionally used in MI risk assessment, a polygenic risk score for coronary artery disease, and retinal vascular complexity measures from both eyes (named fractal dimension). |
| Type Of Material | Computer model/algorithm |
| Year Produced | 2021 |
| Provided To Others? | Yes |
| Impact | This model was evaluated by comparing its internal cross validation performance with the performance of an established model, SCORE. It is observed that our model outperforms SCORE, achieving a higher precision and recall. This implies that the introduction of a retinal vascular trait and a polygenic risk score improve high MI risk population in UK Biobank. This is of great interest as it discloses a promising holistic strategy that can prevent MI incidence and triage those with an elevated MI hazard |
| URL | https://www.medrxiv.org/content/10.1101/2021.12.16.21267446v1.full |
| Title | Stroke risk stratification in UK Biobank |
| Description | A predictive model for stroke (STR) was developed using 455 UKB participants that had an incident event and had a good quality fundus image and an equal number of participants without STR and associated risk factors. This model included clinical variables traditionally used in STR risk assessment, a polygenic risk score for STR, and retinal vascular complexity measures from both eyes (named fractal dimension). |
| Type Of Material | Computer model/algorithm |
| Year Produced | 2022 |
| Provided To Others? | No |
| Impact | This model was evaluated by comparing its internal cross validation performance with the performance of an established model. It is observed that our model outperforms the baseline, achieving a higher precision and recall. This implies that the introduction of a retinal vascular trait and a polygenic risk score improve high STR risk population in UK Biobank. This is of great interest as it discloses a promising holistic strategy that can prevent STR incidence and triage those with an elevated STR hazard |
| Description | Collaboration to complete a meta analysis of retinal vascular complexity |
| Organisation | McMaster University |
| Country | Canada |
| Sector | Academic/University |
| PI Contribution | The main aim of this collaboration is to gain statistical power to elucidate the genetic component of retinal vascular complexity. To this end, each university will complete a standardized retinal vascular complexity genome-wide association study with the dataset each currently uses. We will carry on the standardized GWAS with UK Biobank dataset and then we will complete the meta-analysis with all GWAS summary statistics. |
| Collaborator Contribution | Each of the involved universities will carry on a standardized GWAS. After the meta-analysis is completed, they will carry on functional, enrichment and Mendelian randomization analysis with these. |
| Impact | There is no outcome yet, but a paper is expected to be written with the results. |
| Start Year | 2022 |
| Description | Collaboration to complete a meta analysis of retinal vascular complexity |
| Organisation | University of Dundee |
| Country | United Kingdom |
| Sector | Academic/University |
| PI Contribution | The main aim of this collaboration is to gain statistical power to elucidate the genetic component of retinal vascular complexity. To this end, each university will complete a standardized retinal vascular complexity genome-wide association study with the dataset each currently uses. We will carry on the standardized GWAS with UK Biobank dataset and then we will complete the meta-analysis with all GWAS summary statistics. |
| Collaborator Contribution | Each of the involved universities will carry on a standardized GWAS. After the meta-analysis is completed, they will carry on functional, enrichment and Mendelian randomization analysis with these. |
| Impact | There is no outcome yet, but a paper is expected to be written with the results. |
| Start Year | 2022 |
| Description | Collaboration to complete a meta analysis of retinal vascular complexity |
| Organisation | University of Leeds |
| Country | United Kingdom |
| Sector | Academic/University |
| PI Contribution | The main aim of this collaboration is to gain statistical power to elucidate the genetic component of retinal vascular complexity. To this end, each university will complete a standardized retinal vascular complexity genome-wide association study with the dataset each currently uses. We will carry on the standardized GWAS with UK Biobank dataset and then we will complete the meta-analysis with all GWAS summary statistics. |
| Collaborator Contribution | Each of the involved universities will carry on a standardized GWAS. After the meta-analysis is completed, they will carry on functional, enrichment and Mendelian randomization analysis with these. |
| Impact | There is no outcome yet, but a paper is expected to be written with the results. |
| Start Year | 2022 |
| Description | Eye color Collaboration |
| Organisation | University of Toronto |
| Country | Canada |
| Sector | Academic/University |
| PI Contribution | We derived a Poligenic Risk Score using the GWAS summary statistics for multiple eye colors to assess the influence of eye pigmentation on vascular retinal measurements. The GWAS summary statistics were obtained by Esteban Parra and Frida Durazno. |
| Collaborator Contribution | Our collaborators shared data of multiple GWAS summary statistics which were completed using the data of their project. |
| Impact | This collaboration is mono-disciplinary as both groups are based on studying the genetic architecture of complex human traits. The main outcome of this collaboration was using summary statistics from a recent study to fullfill no-recorded information of the UK Biobank participants. |
| Start Year | 2020 |
| Description | August newsletter entry for British Heart Fundation |
| Form Of Engagement Activity | A magazine, newsletter or online publication |
| Part Of Official Scheme? | No |
| Geographic Reach | National |
| Primary Audience | Other audiences |
| Results and Impact | The application study of the project raised the interest of cardiovascular researchers, which resulted in writing sections to further discuss the potential of the project. |
| Year(s) Of Engagement Activity | 2022 |
| URL | https://sway.office.com/Axst5VQDgx4shWeF?ref=email |
| Description | Centre for medical informatics forum talk |
| Form Of Engagement Activity | A talk or presentation |
| Part Of Official Scheme? | No |
| Geographic Reach | Regional |
| Primary Audience | Postgraduate students |
| Results and Impact | This talk takes place in the university of Edinburgh and it aims to gather all of the postgraduate students in the center of medical informatics and present the project |
| Year(s) Of Engagement Activity | 2022 |
| Description | Centre of Medical Informatics seminar |
| Form Of Engagement Activity | A talk or presentation |
| Part Of Official Scheme? | No |
| Geographic Reach | Local |
| Primary Audience | Other audiences |
| Results and Impact | This seminar is focused to those reseachers within the Centre of medical informatics in the University of Edinburgh. This meeting is focused on presenting all fields of work within the centre as well as collaborators in the centre. This project was presented, and the attendants expressed ongoing interest in this project as there were inquiries about our following steps, managing the database and our limitations and the scope of the project. This meeting helped us understand and develop the characterisation of the project limitations. |
| Year(s) Of Engagement Activity | 2020 |
| Description | DTP Precision Medicine Events |
| Form Of Engagement Activity | Participation in an activity, workshop or similar |
| Part Of Official Scheme? | No |
| Geographic Reach | Regional |
| Primary Audience | Postgraduate students |
| Results and Impact | These events are focused to report ongoing research of those postgraduate students within the DTP Precision Medicine. This project has been presented multiple times to this audience in talks and with poster presentations. Attendants have been interested in this project, as its data management process, methodology and applications have been discussed afterwards. |
| Year(s) Of Engagement Activity | 2019,2020 |
| Description | Eye hospital investors talk |
| Form Of Engagement Activity | A talk or presentation |
| Part Of Official Scheme? | No |
| Geographic Reach | National |
| Primary Audience | Supporters |
| Results and Impact | This talk was focused to report ongoing ocular research to external investors of the new Eye Hospital in the Royal Infirmary of Edinburgh. This project was presented to over 10 people which included the investors and researchers within the field. Attendants reported interest in this project as there were questions about the applications of the projects and its interest. |
| Year(s) Of Engagement Activity | 2019 |
| Description | Eye meeting |
| Form Of Engagement Activity | A talk or presentation |
| Part Of Official Scheme? | No |
| Geographic Reach | Local |
| Primary Audience | Professional Practitioners |
| Results and Impact | This talk is focused to those reseachers, clinicians and industry people working with ocular images. This meeting is focused on presenting all fields of work in the matter of interest and discuss about its application and improvement. The attendants expressed interest in this project as there were inquiries about our following steps, managing the database and its application. |
| Year(s) Of Engagement Activity | 2019,2020 |
| Description | Oral presentation at ESHG |
| Form Of Engagement Activity | A talk or presentation |
| Part Of Official Scheme? | No |
| Geographic Reach | International |
| Primary Audience | Other audiences |
| Results and Impact | This talk was part of the ESHG conference 2022 in Viena. Over 60 individuals attended the in person session which was also available for the remaining attendants. |
| Year(s) Of Engagement Activity | 2022 |
| Description | Poster presentation at ARVO |
| Form Of Engagement Activity | A talk or presentation |
| Part Of Official Scheme? | No |
| Geographic Reach | International |
| Primary Audience | Other audiences |
| Results and Impact | Over 20 clinicians attended to the in-person poster presentation at ARVO session. Further interested ARVO audiences reached out for further details on line. |
| Year(s) Of Engagement Activity | 2022 |
| Description | Poster presentation at UKEGG |
| Form Of Engagement Activity | A talk or presentation |
| Part Of Official Scheme? | No |
| Geographic Reach | National |
| Primary Audience | Professional Practitioners |
| Results and Impact | Over 200 clinicians, academics, and students attended the annual UKEGG conference, where I presented the work from this award. This sparked discussion afterwards and increased the interest of further analysis. |
| Year(s) Of Engagement Activity | 2022 |
| Description | Press release in The Guardian |
| Form Of Engagement Activity | A press release, press conference or response to a media enquiry/interview |
| Part Of Official Scheme? | No |
| Geographic Reach | International |
| Primary Audience | Media (as a channel to the public) |
| Results and Impact | The application study of this project raised the interest of global media. This sparked the discussion of the projects idea and its potential use in clinical settings. |
| Year(s) Of Engagement Activity | 2022 |
| URL | https://www.theguardian.com/society/2022/jun/12/simple-eye-examination-could-predict-heart-attack-ri... |
| Description | UK Biobank eye and vision consortium |
| Form Of Engagement Activity | Participation in an activity, workshop or similar |
| Part Of Official Scheme? | No |
| Geographic Reach | National |
| Primary Audience | Professional Practitioners |
| Results and Impact | This event was focused on reporting ongoing and innovative ocular research using the UK Biobank database and to promote the involvement of charities in this field. Over 60 people attended to this event, which included researchers from different fields, clinicians, ocular cameras developers, UK Biobanks data managers and charities representatives. This project was presented as an application of the VAMPIRE software for fundus images. Attendants reported interest in this project as there were questions about further steps and its applications. |
| Year(s) Of Engagement Activity | 2020 |
| URL | https://www.ukbiobankeyeconsortium.org.uk/ |
| Description | VAMPIRE software meetings |
| 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 | Over 20 people attend regularly to this meeting from different research organizations which are users or developers of the VAMPIRE software. In this meeting, the project normally is presented by me which lead into a discussion about arising issues, further steps, future collaborations, and our research scope. The majority of the attendants reported an increasing interest in the project and suggested ideas to further understand and promote the application the work. |
| Year(s) Of Engagement Activity | 2018,2019,2020,2021 |
| URL | https://vampire.computing.dundee.ac.uk/ |