Make genomic information increasingly easy to save, analyze and interpret. This is the challenge of the engineer Idoia Ochoa, whose research focuses on developing digital formats and algorithms that make personalized medicine a reality.
In 2009, Ochoa left Tecnun, the School of Engineering of the University of Navarra, behind to head to the United States with a grant from LaCaixa. There he completed a master’s degree and a Ph.D. at Stanford UniversityShe went through companies like Google and worked as a professor at the University of Illinois.
In 2020 she returned to Tecnun, this time as a teacher, to continue her research work and share her knowledge and experience with new generations of students. We talked to her about all the possibilities that genome data offers to advance personalized medicine and fight diseases like cancer.
– What does genome data tell about patients?
Genomes can give us a lot of information about diseases, both those that we can suffer and those that we already have. On the one hand, the study of the DNA of healthy cells allows us to discover if there is a predisposition to develop certain diseases or problems. This information can be consulted at any time to work on prevention issues.
In addition, the diseased cell genome gives us other clues. Studying how the genome of cancer cells has changed in a given case helps to define which treatments may work best and to assess whether the disease is more or less aggressive, for example.
– All these data offered by genomes must be processed and analyzed. That’s where your research comes in.
Effectively. With my work I try to make working with genomic data easier and easier. We develop tools to store them efficiently and taking up as little space as possible. We also facilitate that during investigations only certain data can be accessed, without having to download the entire file, and make annotations in the file itself without having to have additional documents.
Another branch of my work focuses on improving data analysis. We develop tools for software and algorithms that allow doctors, biologists and researchers to perform certain actions that were not possible until now.
– What are the benefits of these solutions?
Having efficient tools for storing and managing data saves time for the researcher. It makes it possible to locate, download and share them faster, for example. In addition, the development of new algorithms allows other types of analysis to be carried out and to improve those that already exist.
One of the last articles that we have published in ‘Nature Communications’ refers to Moss, a new tool with which it is possible to analyze several different samples of genomic data from a tumor together. This offers a lot of information and a more complete picture of the disease. Until now, all samples had to be analyzed separately. As a consequence, there was a gap when performing this type of analysis, because it is very expensive to sequence a tumor many times.
– The ultimate goal of all this is to make a personalized medicine. What are its advantages?
First of all, personalized medicine helps the patient. In cancer treatments, knowing the variations of the genome allows doctors to directly give the treatment that will go best. You never know one hundred percent that it will work, but it increases the chances.
“In cancer treatments, knowing the variations of the genome allows doctors to directly give the treatment that will go best”
What has been done so far is to try different treatments until finding the right one, without taking into account all these clues that genomes offer. This wastes money, resources and precious time for the patient.
– When can we have a personalized medicine like the one you advance with your tools?
It depends. Each disease must be studied separately, because they are different and have their singularities. There are diseases in which personalization proceeds quickly and others in which the process is slower because there is less money, less research, or less interest. Breast cancer, for example, is in the first group. It is a disease that has been studied a lot. There are already tests that allow DNA sequencing and determine which people are more predisposed to developing it.
Therefore, I would not say when we can have personalized medicine, but it is already a reality for some diseases. Today the information of the genome is already used to make decisions and, as progress continues, it will be more and more present in our day to day.
– Another objective of your research is that the information can be shared. Are we moving towards a more collaborative type of medicine?
In general, it tends to be shared to the maximum, and this is important because to advance in many areas what is needed is data. Without them you cannot check if your ideas work and demonstrate them later.
“To move forward in many areas, what you need is data. Without them you can’t check if your ideas work and demonstrate them later ”.
Except in exceptional cases, the tools are available free of charge to facilitate investigations. Many open databases are being generated, and also private ones, but to which they allow you to access if you need the information to investigate.
– Projects like Moss facilitate early diagnosis. Will medicine increasingly focus on improving and speeding up diagnoses?
Several factors come into play here. The first, prevention. Knowing that you are prone to certain diseases guides you to change your habits and take care of yourself. Once you have developed a disease, early detection becomes important. Analysis of genomic data can help detect problems much faster.
Lastly, there is the treatment. When it has been verified that someone has a disease, the following question arises: can we use the information from the genome to give a more personalized treatment to this person? In many cases, the answer will be yes. And this is very positive. Above all, for the patient, although it also saves financial resources and makes health systems more efficient.
– What role does technology play in all this?
It’s fundamental. For data analysis in general and genome study in particular, technology and algorithm development are what makes the breakthrough possible. Without them, it would be unthinkable to analyze such large amounts of information
For the development of some of our tools we rely on machine learning and deep learning. Others like Moss are based on theoretical developments and more probabilistic methods, which we later implement in a software.
– You have exchanged the United States for Spain. How do you see the present and the future of research in our country?
I think that more and more research is being invested in Spain, especially in communities such as the Basque Country and Catalonia, but there are still many differences with the United States. One of the obstacles that we find when returning to Spain after doing the doctorate and working as professors abroad is the scarce job offer. In the public university it is difficult to find a position, there are not many scholarships and there are not many offers in the private sector.
This is a big difference from the United States. When I decided to do my doctorate in Spain, everyone asked me if I wanted to be a teacher. It seemed there were no other exits. At Stanford I found that few researchers stayed at the university. They went to big companies like Google or Facebook or started their own companies.
“In Spain there are very good researchers, but it is necessary to improve the interlocking between doctorates, companies and research”
Here, academic training and PhD experience are not so appreciated outside of academia. In Spain there are very good researchers, but there is still need to improve the interlocking between doctorates, companies and research.
Another important difference with the United States is that there, researchers are encouraged to combine the doctorate with internships in companies. I was a summer in Google and another in a start-up. This is beneficial for the doctoral student, because it allows him to test in which fields he prefers to work. And for companies, they have the possibility of meeting potential employees. Many times these practices end with job offers on the table.
– Another aspect to improve is that of conciliation. In your experience, is it possible?
I can send a positive message and say yes. I have three young daughters who were born in the United States and with whom I have now returned to Spain. I believe that it is possible to be a mother, work and be a leader in what you do if the necessary facilities are given.
In this sense, COVID-19 has helped teleworking spread and there is more flexibility. This is another difference with the United States, where, at least in my sector, work is measured by results and there is not so much obsession with schedules as long as the objectives are met. I think this is essential so that families and especially women can have the career they want.
In Nobbot | The Human Genome Project revolution turns two decades
Images | Idoia Ochoa, Unsplash / National Cancer Institute, Unsplash / Yassube Khalfalli