‘If we don’t have the skills to interpret the data we won’t be able to understand what is happening to the patient’

My role is to help interpret people’s DNA together with symptoms and laboratory tests to identify the cause of disease.

How long have you worked here?
I’ve been based in Cambridge for over four years. Previously, I worked at the Cancer Research UK Cambridge Institute as a postdoctoral researcher. My work there also involved genetic data but the research was less clinical.

What does your role involve?

When a blood or saliva sample is taken, it is processed to obtain genetic data and then it can be analysed. Most of my time is spent trying to understand the relationship between the genetic and the clinical data that we collect. Each human genome comprises 3 billion letters, and we have to consider slight differences in these letters between individuals to try to identify the cause of the disease.

Much of our research focuses on people with rare diseases. These people are grouped together broadly according to general features of their disease but when we look at their symptoms and test results in detail they become quite distinct. We collect the data to try to work out whether people with similar problems have similar changes to their genomes to try to determine the underlying cause of disease.

What is a genome?

A genome can be thought of as a set of very long DNA sequences totaling 3 billion letters in length. We have two sets of these letters, with one set coming from one’s mother and the other from one’s father. Parents pass on roughly half of their genomic DNA to their offspring. The precise “half” that a parent passes on is random, and this is the main reason why each individual is genetically unique, with the exception of ‘identical’ twins.

A human genome includes about 20,000 genes, which are present in almost every cell in the body. Different cells use different sets of genes, which is one of the main reasons why cells are so diverse. If a gene has a defect impairing its function, then cells in which that gene plays an important role may not function normally.

What does it mean to have a rare disease?

There are probably about 8,000 different hereditary, severe rare diseases. The genetic basis of roughly half of those is known. Each one is individually rare but as there are so many rare diseases, it is estimated they may affect 1 in 17 people, which equates to around 30 million people across Europe. However, two people may have the same disease and similar symptoms, but when you go back and look at the patient’s molecular results, there may be different underlying reasons for their conditions. Most people have heard of the bleeding disorder haemophilia. It could come under the bracket of rare disease even though a lot of people know about it. A rare disease is not necessarily something you have never heard of or is brand new.

Why were you interested in this area of research?

What I like most about it is that it spans many aspects of science. My work combines statistics, computer science, biology and medicine, which I find very interesting.

However, my academic background is in computer science. I completed a PhD in biostatistics and I learnt the biology as I went along over the past 10 years. My background is quite mixed, which is a good match for this kind of work. I didn’t expect to go into a clinical environment but I have learnt a lot about some aspects of medicine by working with medical colleagues.

Why is your role important to research?

In the past medical and lab tests were expensive and therefore they had to be done sparingly, based on judgement by a doctor. Over time many tests have become much cheaper. For example, the test to count the different types of cells in the blood is simple and cheap. A machine takes tens of thousands of measurements in one go, how many platelets per millilitre, their size, shapes, etc. Currently, the results of the test are reduced to 12 values, which are used in the clinic to reach a diagnosis. There is, however, a lot of additional information hidden in these simple measurements and that’s where people like me come in. We use these routinely generated data from NHS tests to try to get a better understanding of the link between disease and genetic alterations.

For example, a rare disease of the blood originates from a small number of blood stem cells. Blood stem cells divide into different cells, which multiply and spread to replace the blood cells at the end of their lifespan. To understand this process of blood cell formation, mathematical models and statistical methods need to be developed. It turns out that many medical and biological problems require statistical analysis and that’s why I have a role to play with the medical community, biologists and geneticists.

How do you look at data?

An interesting aspect of rare diseases is that they typically span multiple organ systems. If someone has poor hearing you can work that out with a relatively simple instrument, but the precise type of impaired hearing may only be determined with more specialised equipment. People with the same disease are enrolled in many different clinical centres across the world and the referring clinicians will use slightly different approaches and record data differently. They could record ‘poor hearing’, ‘hearing impairment’, ‘sensorineural hearing loss’ or even ‘high-frequency sensorineural hearing loss ’, for example.

We are now starting to code this detailed information in databases and are using a universal language to describe patient symptoms. The terms describing these symptoms and the relationships between them are recorded for each patient in a graph. We then search for people who have similar graphs/symptoms, who also share similar genetic changes to find groups of individuals who may have the same rare disease. This can be quite challenging because different gene defects can cause the same disease and the same gene defect can manifest differently in different patients. One person might have kidney failure and cataracts and another person with the same gene defect might not have either of these conditions.

A clinician wouldn’t have time or the ability to go through all the information in thousands of individuals in one go and that’s why our jobs exist to make sense of it in an automated, computerised way.

How does this language work across the world?

This language for recording patient symptoms is being used by the 100,000 Genomes project, which is happening nation-wide, amongst many others. There are also similar, but smaller projects in America and Continental Europe. Recording the clinical data in this way and using the same standards allows us to share data to increase the chance of finding the genetic causes underlying the rare disorders.

What kinds of people work with you in research?

There are many different people with different expertise that help the research. There are probably 100 people involved to help collect and understand the data: people in IT, laboratory technicians, administrators, biologists, clinicians, nurses, geneticists, bioinformaticians, biostatisticians and others. A diverse set of people come together to retrieve the samples, process them and then help with the interpretation. We are very lucky to have committed staff in all these areas who want to contribute to our understanding of disease and how genes function. We have regular contact with many medical consultants without whom we wouldn’t be able to move forward with this kind of research.

How long does it take you to analyse the data so new treatments can be made?

It all depends on what we’re looking for. A recent study I worked on found a mutation in a well-studied gene that caused a number of symptoms. Two years ago clinicians couldn’t understand why this particular gene was causing problems. We looked at the data from test results and we were able to find the cause and think about potential treatments. In this case the disease was due to an alteration of a gene leading to an over-active protein. There were already approved drugs on the market to help inhibit the activity of the protein coded by this gene.

There will be many cases that will remain unexplained because they involve highly complicated biological mechanisms. For example, it may be that a number of different genes are not working together as they should. There may be a significant delay for new treatments but our incremental understanding will likely bring benefits further into the future. In some cases the work researchers are doing today may only have a tangible impact for patients 20-30 years into the future. However, not everything is years away and some existing drugs need to be re-purposed and tested to see whether they also work for another disease.

How do volunteers help you with your research?

The main volunteers are the patients with rare diseases and their close relatives, as they obviously play an essential role in our research, often without expectation of personal gain. Without them we wouldn’t be able to extract data from their samples to search for the root cause of disease or to help us understand it.

Many of the clinicians are in a sense volunteers because they don’t have to work in research in addition to doing their day job, which is looking after patients. Many clinicians participate because they want to make a difference to health care in the future. All staff members involved in research are hugely important to help us with our work.

There are a large number of blood donors of NHS Blood and Transplant (NHSBT) that play a significant part in research as well. When people donate blood they can consent for their samples to be used in research. We can then have access to people who may have specific traits or genetic characteristics that we wish to investigate in the course of a rare disease project, even if they themselves do not have a rare disease.

Why should people take part in research?

I have the impression that many people with rare diseases want to know what the cause is so they may better understand their disease, irrespective of practical implications. I believe there is also an altruistic component to their motivations. People with a rare disease are uniquely empowered to make a contribution to research as, there may only be a handful of individuals with their condition in the entire world.

By giving a sample and answering some questions, these individuals can help move science forward and potentially help others in future. Where there is a family history of a disease, people are motivated to take part in order to prevent passing it on to future family members. Finally, research can also be a route into a community of individuals who have similar disorders, allowing patients to connect with each other and share experiences.

How much do you tell the public about your research findings?

Everything we do is funded through the government, research councils and charities. Our results are published, usually in open access journals, which means anyone can go online and read about the findings. Our work is paid for by the public and we try to ensure our output is available to everyone for free.

How do you protect the data of people who take part in clinical trials?

It is entirely reasonable and legitimate to be concerned about privacy issues. We take all the steps necessary to protect confidential data. I have access to detailed patient data, such as blood counts, whether they have autism, etc. but I don’t know their actual identity, their name or where they live. All the data are anonymised and we are not allowed to remove certain kinds of data from particular computer servers. Our policies are such that we can do our work but with a very low risk of any security issues arising.

If people want to get involved in research, who should they contact?

The National Institute for Health Research (NIHR) Cambridge BioResource and NHS Blood and Transplant centre are good places to get involved in research as a healthy volunteer or as a patient. I also suggest people discuss the experience with others who have already taken part in research.

What do you do in your spare time?

Research doesn’t leave much room for recreation these days, but when time allows, I enjoy playing and writing music. I’d like to record a short collection of lullabies for my nephew, who is due to be born soon. My main concern would be completing it while he’s still of an appropriate age!

What would be your dream job?

I’m very fortunate to be working with great colleagues and collaborators. I would like to have my own independent research group and continue to produce high quality research in a well-funded environment.