In Computational Combat Against Microbial Resistance
Victoria Carr on gene sequencing, women in tech and getting out there.
Victoria Carr is a Bioinformatics and Data Science PhD Researcher at King’s College London. She is currently investigating antimicrobial resistance and developing a risk score method from sequenced metagenomes. She is the founder of Researc/hers Code, a community showcasing the talents of women and individuals of diverse backgrounds working on computational research. Previously, she was president for King’s College London at the Innovation Forum.
So, you work in computational genomics. Can you tell us about the field and your research?
Computational genomics is using computer science to understand more about living things from DNA data. My research looks into discovering antimicrobial resistance from metagenomes. A metagenome is all the DNA that can be found in a sample. In my case, I analyse the antimicrobial resistance in the metagenomes of all the microbes that live in the human oral cavity and gut.
Why is microbial resistance (and preventing it) so important?
Antimicrobial resistance is a threat to global health and the environment. A high profile report predicts by 2050 10 million people worldwide will die of antimicrobial resistance infectious per year. By 2050, you may be more likely to die of an infection that is easily treatable now, than to get cancer. But this could be prevented if scientific institutions, industry and governments work together and act now. Already, we’ve seen a huge amount of investment in antimicrobial resistance research from the UK government and accelerators, like the revival of the Longitude Prize.
What creates antimicrobial resistance and can it spread?
The biggest cause of resistance to antimicrobial drugs is the overuse and mis-use of these drugs themselves, both in healthcare and agriculture. After a course of an antimicrobial drug, some microbes that are stronger survive. They may then multiply to create a strain that is more resistant to the drug. Individuals worldwide are often prescribed antimicrobials unnecessarily and do not take these drugs correctly. In hospitals, antimicrobial drugs are used routinely in surgery, making hospitals a potential incubator of super-bugs, such as MRSA. In agriculture, animals are constantly administered antibiotics to make them bigger and fatter (rather than to treat infections) and especially in arable farming, fungicides are routinely applied on crops. Strains that carry resistance can spread from human to human, animal to animal and between humans and animals. For countries and communities where antimicrobials are misused in healthcare and agriculture, antimicrobial resistance infections are more likely to be passed to individuals who live in overcrowded living conditions with poor sanitation, and develop in those that have poor access to healthcare.But this isn’t the only way resistance can spread. Antimicrobial resistance genes can hop from one microbe to another microbe. This means resistance genes found in harmless, friendly microbes can cross to harmful microbes. This is exactly how the superbug MRSA developed, which killed hundreds of people in the UK and other Western countries. It picked up a gene that was resistant to the antibiotic, methicillin, from a harmless species. These resistance genes are stuck to “mobile genetic elements” that hop across species via a process called “horizontal gene transfer”. This process is very common in nature, which is why the spread of antimicrobial resistance is such a big threat. But not much is known about when, where and how it works.
” Antimicrobial resistance genes can hop from one microbe to another microbe. This means resistance genes found in harmless, friendly microbes can cross to harmful microbes”
Which genes are you trying to find?
Because of the way resistance genes cross between microbes, I am particularly interested in finding mobile genetic elements as well as resistance genes from metagenomes. Another trait is mobile genetic elements tend to cross between similarly related species, so I am also having a look into the types of species that exist. The ultimate goal would be to create an antimicrobial resistance “risk” score for an antibiotic, by tying together what antimicrobial resistance genes are there, what mobile genetic elements and species exist and whether they are linked to these resistance genes. Predicting the onset of new resistant infections in this way could help target policies to communities at higher risk of resistance before an epidemic strikes. For example, regulating the use of an antibiotic where there is a high resistance risk for that antibiotic.
Who will the risk score help most?
A risk predictor would most likely help individuals in communities who are exposed to a source of antimicrobial resistance, such as patients and staff in hospitals, farmers, and people living in overcrowded conditions. It’s not just people that could benefit, animals could also benefit from this too. I am always on the look out for animal metagenomes to add on to my project!
So, we shouldn’t we stop using antibiotics altogether?
” So much high-resolution microscopy and imaging technology in biology and healthcare rely on specialised software. AI is now becoming used more for image and feature recognition too “
Absolutely not! It’s about whether it is necessary to prescribe antibiotics. In some areas, such as clusters of villages in Ethiopia, antibiotics are difficult to get hold of, and people are dying of treatable infections. But still people are taking antibiotics unnecessarily all over the world. One classic case is taking antibiotics to treat a common cold, which is caused by a viral, not a bacteria infection. Instead of treating the cold, the antibiotics would be completely ineffective but would increase the risk of the individual becoming resistant to that antibiotic. To combat this and give the correct drugs, we need a point-of-care diagnostic that determines whether an infection is bacterial, fungal, viral or parasitic. But the real challenge is that it has to be cheaper than the cost of an antibiotic.
Your research is at the crossroads between Computer Science and Biology, how did you end up on this path and are there other ways to employ computer science in research?
I do not come from traditional computer science degree, but a biology undergraduate degree. I fell in to computer science around about the 2nd to 3rd year. I did an internship in computational biology between in my 2nd and 3rd year at undergrad (partly because I wasn’t having much luck getting wet lab placements!) I realised then that so much bioscience research depends on data analysis. So I jumped on the bandwagon and did a Masters degree in Systems Biology, where I got my formal training. It’s not just ‘omics that benefit from computer science. So much high-resolution microscopy and imaging technology in biology and healthcare rely on specialised software. AI is now becoming used more for image and feature recognition too.
What drove you to start Researc/hers Code?
Before I started Researc/hers Code a year ago, I attended a lot of “women in tech” meet-ups and community events around London. They were always fantastic and I met so many interesting and inspiring people, but I noticed that at a lot of these events there were speakers mainly from industry and not so many from academia. This really surprised me because there are so many incredible tech women in academia, but less widely known in the London tech community. I decided to take the matter into my own hands and start a community for women and minorities who work on computational science to talk about their research to a wider audience.
What are your future plans for Researc/hers Code?
Researc/hers Code has potential to grow in a big way. So far, many of the meet-ups have been hosted at academic institutions, but the next one will be held QuantumBlack, a data analytics firm, which we are very excited about! A website is coming soon and more volunteers are joining us.
” Somebody interested in fashion may like the idea of developing apps for retail companies “
Some of the earliest computer programmers were women, yet that’s not the typical stereotype most people are familiar with. Do you have any ideas of how we can increase the representation of women and people from diverse backgrounds in Computer Science?
There are so many different ways you can help the women get into computer science from an early age to adulthood. For girls and young women who are at an age where they inevitably may be influenced by societal gender biases, it is important to show the variety of things computer science can do depending on their interests. Somebody interested in fashion may like the idea of developing apps for retail companies, or a computer program that could diagnose cancer may appeal to somebody who likes biology. While I was growing up and choosing my GCSEs and A-levels, I never considered the idea of doing computer science. I don’t think computer science was offered as a subject, but it was offered at the local single-sex boys school down the road. My physics teacher (ironically) told me that computer science was for men only. If I had the opportunity to study computer science, then I may have considered a computer science degree. In short, it’s important for schools, institutions and companies to present opportunities that are accessible to everyone.
What advice would you give to other young researchers who might be interested in the field but aren’t quite sure if Computer Science is right?
Go to community meet-ups and events on something you find interesting. Get out there and meet people! It’s daunting at first, but you’ll quickly find friendly faces who will be happy to give advice or share their experiences. If you’re not sure computer science is for you, then definitely drop by at a Researc/hers Code event! You can sign up to our meet-up page.