Crowd sourcing new treatments for malaria
New medicines are urgently needed for diseases of the developing world and those medicines need to be affordable. Scientists sharing their information and collaborating openly rather than keeping their work hidden in a lab could work more effectively to solve medical problems.
That is the idea Mat Todd from the University of Sydney had when he was trying to improve a medicine currently used in the developing world. He thought open source collaboration has led to the development of incredible software products - Linux, Google Chrome and Firefox – why not apply the same model to drug discovery? The work he did with World Health Org on this project took place on the web in real time, and it's now being taken to market.
At the same time, GSK deposited more than 13,500 structures of possible drugs against malaria into the public domain, along with associated pharmacological data. It was the first large-scale public release of such structures by a pharmaceutical company. GSK also created an "open lab" based at the company's neglected-diseases research facility at Tres Cantos in Spain, which offers top international scientists and academics the opportunity to pursue their own projects as part of an integrated team. Researchers are able to access GSK’s expertise, processes, and industrial-scale infrastructure as part of a collaborative approach to drug discovery.
Ass Prof Mat Todd has been involved in two projects at the GSK lab – one in malaria and one in TB.
More about our open innovation approach in diseases of the development world
In our work on the treatment and prevention of malaria, tuberculosis and kinetoplastid infections (African sleeping sickness, Chagas disease and Leishmaniasis), we have used the web to share the data from our investigations into potential new treatments. We’ve screened our entire library of more than two million compounds and shared information on those that show signs of activity.
In May 2010, the first success in opening up access to our compounds was realised, with the publication of over 13,500 promising potential ‘hits’ in the journal Nature to stimulate drug discovery research for malaria. The chemical structures and associated assay data of these compounds are now stored on leading public scientific websites including the European Bioinformatics Institute, the National Library of Medicine and Collaborative Drug Discovery.
This open approach has seen us share our anti-malarial data with 14 research institutions around the world, resulting in a number of new research projects. A pre-requisite for granting access to the data is that the researchers agree to put their findings in to the public domain, thus encouraging further collaborative research by the scientific community on this challenging disease.
The group Medicines for Malaria Ventures (MMV), which partly funded our initial malaria screening, has also been instrumental in coordinating this open source approach. MMV have created a ‘malaria box set’ made up of the compounds donated by us, and other research groups including St Jude’s Children’s Research Hospital and Novartis. The malaria box has been sent to more than 100 groups around the world.
Several labs are also involved in an exciting new open source drug discovery project for malaria - the first of its kind - using a new idea called ‘Open Notebook Science’. This involves publishing the notebook of the researcher online, along with all the raw and processed data and any associated material, as it is recorded. Led by the Todd lab at the University of Sydney with MMV and GSK’s Tres Cantos facility, this new practice hopes to speed up the collaboration process.
Tuberculosis (TB) research
In October 2012 we announced we were adopting the same open approach to TB research, by putting around 200 TB “hits” into the public domain. In the same way that we previously opened up access to our malaria data, we will make these TB data freely available to the public online and will also seek publication of this information in a peer-reviewed scientific journal. We hope the release of these data will encourage a fully open approach to TB research, which we believe is the key to accelerating the development of new medicines to treat this disease.
Kinetoplastids infect an estimated 20 million people in the developing world, resulting in approximately 95,000 deaths a year. Yet despite the enormous suffering they cause, these diseases have historically received a limited amount of attention and effective treatments are lacking. To stimulate research in this area, between 2012 and 2014 we carried out a similar screening exercise for kinetoplastid infections as we had previously done for malaria and tuberculosis. In March 2015 the results of this process were published in a peer-reviewed scientific journal and we have made the c 600 “hits” identified during the screening available to researchers, to encourage further research in this field.