Author: Yumna Moosa - 2017-07-28Tweet
I was doing an all-night shift in the hospital where I'm working this year when I received an e-mail offering a 1,000 USD Travel Fellowship to attend a conference in Eastern Europe. But in order to receive the grant I had to register for the conference (and pay the not insignificant fees) by 9am the next day. Classic 419 scam, right?
Fortunately, it wasn't, and today I find myself in Prague, Czech Republic, buzzing after three days at the world's premier bioinformatics and computational biology meeting: the annual meeting of the International Society for Computational Biology, the 25th Intelligent Systems in Molecular Biology Conference.
I am an MSc candidate at KRISP, but this year I've been a bit out of the scientific loop while completing my clinical community service at a small rural hospital in northern KwaZulu-Natal. In this small seven month small break, it feels like the ground has already shifted significantly. I am amazed at the work being done in the fields of personalised medicine, human-microbiome interactions and human data science. At times it felt a bit like I was an extra in a science fiction movie. I want to share a few highlights.
1. Open Humans
Madeleine Ball is the founder of the Open Humans Foundation, a not-for-profit organisation that seeks to give full control of personal genomic and health data to the individual from whom the data is collected. Currently this sort of data stays with the researchers who keep it safely locked away, inaccessible to the individuals or any scientist who may want to reuse the data.
As could be predicted, the idea of sharing personally identifying data inspired some unease in the audience. Genetic discrimination. Uncovering the existence of a previously unknown relative. In discussing the various ethical, legal and social implications of personal data sharing, one delegate made an interesting point:
'A few decades ago a new technology was developed that could store personally identifying information such as disease risk profile, ethnicity or hidden parentage and there were similar concerns as those we're describing. But today just about everyone has a camera on their phone and shares photographs freely!'
2. Open Access Publishing
The existing commercial scientific publishing model is as follows.
- Public pays Scientist to do science.
- Scientist does science.
- Scientist gives science to commercial Publisher, who does not pay Scientist, but takes over full copyright of science. Scientist may no longer share or use any of this work without Publisher's permission.
- Reviewer, another scientist, reviews science but does not get paid for this service.
- Science is published.
- Public pays Publisher to access Science.
See the problem? The public pays twice. For a more comprehensive argument as to why scientific publishing needs an overhaul, see this piece in the Guardian.
I had the pleasure of chatting to some of the Public Library of Science (PLoS) people at the conference. PLoS is a not-for-profit Open Access publisher, with a different funding model, allowing the public to pay for the science to be published and for the work to then be accessible to anyone with an internet connection. Commercial or Open Access, it is the same scientists do the same work, which is reviewed by the same peers, according to the same standards of scientific rigour.
3. Open Data
Fiona Nielsen started out as a frustrated PhD student, severely constrained by the inaccessibility of human genomic data, and has since transformed into an inspiring entrepreneur championing open data in human genomics. She was so determined that she established both a charity and a company in pursuit of the cause. She speaks about the FAIR principles of Open Data:
Everyone loves a good acronym! She gave a keynote address at the Student Council Symposium and I want to be like her when I grow up.
Here at KRISP we're developing a viral metagenomic pipeline, and access to suitable data for benchmarking remains a nightmare. I just hope that her company Repositive expands its services to include metagenomic data ASAP.
4. Personalised medicine
Christoph Bock delivered a keynote address entitled 'Bioinformatics for Personalized Medicine: Looking Beyond the Genome' where he discussed how the complex function of our bodies requires trillions of individual cells to integrate, interact, and strike the right balance between stability and plasticity. This requires both the basic blueprint of all available possibilities as encoded in our genome, and other layers of regulation that switch on or turn off specific genes under appropriate circumstances. The latter is commonly referred to as the epigenome. With the use of CRISPR-cas9 technology we have the capacity to alter these epigenomic instructions, and thus influence the activity of our own genes. It is really quite fantastical. The best part? My colleague at KRISP, Veron Ramsuran, works precisely in this field. Maybe I didn't have to go all the way to Prague after all...
I came back energised by all of the scientific ideas that we can implement in KRISP when I return, after only five more months of 'saving lives'.
Yumna Moosa is an amateur mathematician who studied medicine by mistake, inspired by a naive desire to 'help people'. After completing her medical internship she came to her senses and registered for an MSc in Virology and Bioinformatics at KRISP, and is working to develop software for virome analysis. She is interested in human-microbiome interactions, and the way that these complex microbiological networks influence health and disease. This is partly because she has New Age tendencies and likes to believe that 'We are all One'. She is currently completing her community service at Bethesda Hospital in rural KwaZulu Natal.