In this video, we discuss how we helped one of our customers, OMass Therapeutics, accelerate their drug discovery process from one year to seven hours.
OMass Therapeutics are an Oxford-based biotechnology company using native mass spectrometry and other biophysical technologies to drive drug discovery in high definition. The company is building a pipeline of small molecule therapeutics to bring life-changing benefits to patients suffering from immunology and genetic disorders.
As everyone probably knows, drug discovery can be quite a long process, involving multi-year programmes and a large investment. What companies aim to do is figure out how the protein or drug they’re planning to release will work.
Ideally, most organisations will do this virtually first, before they move into more expensive, physical testing. One of the initial processes is called virtual screening and involves taking the protein and assessing it virtually against a given set of molecules. The aim is to get as many hits as possible, before moving into the physical world of testing and so forth.
So, the challenge really is – how much can you actually do in that virtual part of the programme? How much can you screen? How cost effectively can you screen? How much compute time can you afford?
The more widely you can do your virtual screening, the greater the likelihood of hits that will hopefully lead to successes later on in the programme.
To use OMass as an example, they were screening about a million molecules over a period of a week. However, what cloud enables them to do is to build up a larger cluster, spread across multiple different regions, using resources which could be up to 60-70% less expensive – for example, using AWS Spot instances.
What this means is, rather than screening a million molecules in a week, companies, such as OMass, can now screen multi-millions in a matter of hours.
So, what we did was effectively create those on-demand clusters for OMass.
We ran just over 378,000 virtual CPUs, or virtual cores, across four geographically dispersed regions, which is a huge amount. Not only was the scale significant, but we also used Spot instances, to make the clusters as cost-efficient as possible.
What this enabled OMass to do was virtually screen about 330 million compounds in the space of 7 hours.
In addition to this acceleration, we provided OMass with a solution that was on-demand. Therefore, if they asked us today, to bring up a similar cluster and do another run, we could do that at the touch of a button.
I think there are two key benefits –
The wider you can screen, the better. Given the virtual element, the cloud gives you a lot more flexibility than on-premise infrastructure, so you have more assurance before you order the compounds and begin the chemical testing.
The deeper you can go effectively, the higher your success rate will be throughout the rest of the programme.
As this process is repeatable, you can apply the same scale to the next programme, the following programme, and so on, without the manual effort of building the clusters each time.
Given the benefits new drugs and therapeutics can bring patients, acceleration of this magnitude is really significant.
To read more about how we helped OMass, read our case study.
You are seeing this because you are using a browser that is not supported. The YellowDog website is built using modern technology and standards. We recommend upgrading your browser with one of the following to properly view our website:Windows
Please note that this is not an exhaustive list of browsers. We also do not intend to recommend a particular manufacturer's browser over another's; only to suggest upgrading to a browser version that is compliant with current standards to give you the best and most secure browsing experience.