Professor Aedín Culhane

Aedín Culhane is Professor of Biomedical Sciences (Cancer Genomics) and Director of the Limerick Digital Cancer Research Centre in UL.

She is a computational oncologist with expertise in multi-omics data integration, statistical genomics, clinical bioinformatics and genomics in oncology.  She has over 20 years experience in cancer genomics, of which over 15 years were in the Dana-Farber Cancer Institute and Harvard TH Chan School of Public Health in Boston, USA.

Her recent research focuses on algorithm development and integrative data of single cell molecular data in cancer to identify molecules that regulate and can be targeted during tumour development, progression, drug response and resistance.  She is a leader in the Bioconductor community, a global open source, open development software in R for genomics and an advocate for open source science.

Professor Aedín Culhane is curious about cancer and how we can diagnose it, how we can prevent the disease, and how we can develop better treatments for cancer.

She’s trying to understand cancer at the single cell level, the real fine detail, trying to work out what exactly has gone wrong with the cancer cell, and also how we can use the cells or normal cells around the cancer cell to actually treat cancer.

Cancer and medicine are now a data science

We are generating enormous amounts of data because the technology, like those microscopes to look at cancer, have got so much more detail down to the single cell level. And so that technology that's generating the labs then generates lots of data.

So, working in cancer harnesses expertise across disciplines from basic biology, bench biology in the lab, pathology, looking at the cells, of course the data science, the computer science, the AI, the machine learning that's required to analyse these vast amounts of learning and vast amounts of data.

And then bringing it the whole way back to the clinicians and the clinic because that's where we want the most impact.

The genome was sequenced about 20 years ago, but the actual final version of that genome wasn't actually available until last year. So, over the past 10, 15 years, we've been developing genomic tools, that's tools that can look at the genome. And these have been progressing and advancing and advancing and advancing. So, the tools we have today are far more detailed than what they were five or 10 years ago. And so, the data that we generate is obviously vastly more, so we can now generate information on every single cell, which is probably tens of thousands of cells in a single piece of tissue, and then on tens of thousands of genes. So, these are very, very large matrices of information and we need computers to analyse that information.

Delving into the unknown

The field that we work in, clinical cancer, computational biology is brand new. It didn't exist five or ten years ago. So, we're really developing a brand-new field because today we can take clinical samples and we can work with clinicians to identify cohorts of patient's forest study, and we can provide molecular data at such fine detail that we never could before, that we can really begin to interrogate the disease. That was impossible five years ago. That requires advanced mathematics, biostatistics, computational biology and machine learning in AI.

We're really bringing together many, many different fields to tackle this really important question. We work with people that come from a physics background. We work with biologists, we work with mathematicians with biostatisticians.

There are a number of people that come from various numerical fields to tackle this challenge. It creates a really vibrant research culture because we're experiencing so many different ideas together. And it really is that exciting field when you just get so many people that have so many different ideas together.

For cancer, computational biology the big things have been the sequencing of the human genome and also the massive developments in computers in the internet, how we share data, how we analyse data. Without those two coming together, we wouldn't be doing this field. And if you think yourself of the changes that you've observed within the past five years in computers, going from paper maps to Google maps. Things that people relied on, 15-20 years ago are now defunct.

But cancer research has undergone that same change. So, the way that we do things now is completely different to how we did it ten or fifteen years ago because we have new tools because of the sequencing of the human genome and because of that we are using completely new ways to look at cancer and we need really bright, smart, motivated people to work with us because this is an enormous challenge. 

Every single day there's another new paper that somebody has published in the field that changes how we think. So, you have to be ready to adapt. You have to be ready to talk to a lot of people because we don't work in silos. It's not like a group working in Limerick on their own. It's not even a group is working in Ireland on their own. We're immersed in a global community. So, it's important to like people and to like working with people and it's really important to be able to delve deep into a subject, understand it and advance the field.

Targeted approach for patient centred healthcare

In the 1980s and 1990s, most people got treated with chemotherapy and radiation. Chemotherapy and radiation target the fast-growing cells, like literally any fast-growing cells, cells that are dividing. That means that when you target cancer, we also lost our hair because our hair is growing. We targeted the immune system and there were a lot of side effects. There's a tonne of side effects from chemotherapy cause we're targeting not just the cancer cell but also the normal cells.

Now, can we actually design drugs that target the cancer specifically because then you won't have all of the side effects. And those side effects are so important to patients who are undergoing treatment because if the treatment makes you really sick, then you don't have the quality of life to spend with your family.

If we can give patients targeted therapy that's just targeting the cancer cells, then people can actually have a pretty normal good high quality of life with their families. And that's, that's what this is all about. It's all about impacting patients and their families.

In the past ten years there's been a lot of development of drugs that actually target specific mutations in the genes. And then more recently we're trying to use the body's own immune system to target cancer. So, as we learn more and more about the disease and also how it interacts with all the cells around it, we're developing more and more targeted therapies.

And as we develop more targeted therapies, we actually first of all do a better job of tacking the cancer cells. And secondly, patients have fewer side effects. Person-centred healthcare is our ability to take that disease and understand it in your context.

So, what is your disease? And then develop a medicine which will actually target your disease rather than using something like a broad chemotherapy and giving everybody the same treatment. So, the biggest impact that we can have for patients is if we increase the number of clinical trials that are available to patients. When patients are getting treated for cancer, they will have like the first line, the second line of therapy. And then at that point in time they're often looking for the new medicines, the thing that might actually treat their cancer.

We have connected into seven different cancer centres in Ireland. And what we are doing is working directly with clinicians, trying to see if we can harmonise the data. If we can get a harmonised local copy of their data and that data is in the same format as each hospital, we can now send a query to each one of those sites to then get statistics on the number of patients that are at each clinic, their genomics, what they're being treated for, and we can begin to get a better national picture of cancer in Ireland.

They can now discover the nuances in the genome that are important for local diseases because the diseases that are important in western society are not the same diseases that are important in Sub-Saharan Africa where disease and infection is so important and where they need to understand the local genomes and the diversity of that genomes within Bioconductor.

We provide them the tools to actually analyse that genomic data. And when we asked our global community what they needed, they said training. And so that's what we're leading from our project here in UL is actually a global training for Bioconductor.

Open-source science is fundamental to the values in my lab and, and in what we do. We build on the knowledge of others, but also, we contribute to the global knowledge that's available and it's how we can all work together, not compete with each other, how we can work together to solve these bigger challenges like cancer. We are really just like a tiny pin in an enormous chain and everybody just adds to this effort. 

From the local to the global

We need to move away from thinking of, this is Limerick versus Dublin versus Cork or any of this kind of local competitiveness. We really are tackling a much bigger challenge [in cancer] and we need to collaborate within Ireland, across the border. We need to collaborate within Europe, and we need to collaborate within the US.

We all need to work together to solve these problems. And that's what open source does. We all work with each other, we all share our code, we share our data and we build on the knowledge of others. The worst thing I think that puts so many people off science is they think it's only for like these incredibly smart people. Like, the Einsteins, the guy who jumps, "Eureka!" from the bath or something and that's not how science works at all.

It's much more methodical and it's much more iterative. It's building on what the previous person did. And we take it a little step forward and a little step forward and a little step forward!

And the importance is doing that each step. We're climbing a stairs, it's a really big stairs and we're taking it one step at a time. Nobody's leaping to the top. But actually, the person that sits there and just can actually do that work is actually just so important that we just need people who are engaged, who are motivated.

The Limerick Digital Cancer Research Centre (LDCRC) aims to use the newest, computer science, maths, statistics, our technology sensors to actually understand cancer. 

It's how we can leverage strengths of UL, our strengths in development of nanomaterials, our strengths in biomaterials, our strengths in sensors, and then our strengths in data analysis, in computer science, in maths, and in stats to actually analyse the data in cancer and cancer data which has changed entirely in the past five to 10 years.

We now have the technology that actually can fine tune, understand cancers at the individual single cell level. And that generates vast amounts of data and vast amounts of data need maths, stats, computer science and AI to understand them. And by leveraging that strength of UL, we are advancing our understanding of cancer. 

There's two major focuses of the LDCRC. One is in pathology - understanding the cancer on the slide, the tissue, and looking at that at the single cell level. Looking at every individual cell in the cancer and also all the normal cells surrounding it. The second is molecularly profiling, look at all the genes and proteins that are actually in that cancer. And that generates vast amounts of data for which we need the maths, the stats, and the AI.

So that second component is really the bioinformatics, the cancer computational biology, and it's how we can bring the best minds to actually analyse that data to develop cancer data science.

The LDCRC here in UL has a mission to understand cancer. It is using the strengths of UL in cancer research, strengths in data science and then also the tools that we use to interrogate cancer. So, we're using digital pathology, we're using sensors, we're using new technology so that we can understand the makeup of cancer and its detail at the molecular level.

Curious about Aedín Culhane's research