← Back to News

8x faster COVID-19 testing by pooling samples

Mon Nov 04 2019

www.pcr-pooling.com provides a tool to support COVID-19 testing laboratories to increase their throughput of rtPCR testing by up to 8X without the requirement of new equipment or training for new skills.

biotx.ai's COVID-19 strategy is to find patterns linked to the disease in the genomes of patients as well as in the DNA of the virus itself. This process takes time, and the limiting factor is the rate at which the data becomes available.

All the better that our CTO, Charles Ravarani, has developed a fast and innovative solution to one of the biggest COVID-19 challenges: Mass testing.
Using simple combinatorics, a branch of mathematics that studies properties of finite collections, Charles' method speeds up mass testing for COVID-19 by the factor 8 (!). The solution is ready NOW:

Have a look at www.pcr-pooling.com

More details by Charles himself below:

The current global COVID-19 health crisis is affecting all of us in some way or another. Many countries have imposed social distancing guidelines or imposed full-scale lockdowns with the aim to slow the spread of the disease. The idea is to reduce social interactions in order to lower the risk that infected, potentially asymptomatic individuals will infect others. However, locking down everyone, including healthy individuals, to be on the safe side comes at great cost to our normal way of life.

Some countries with a well-developed testing infrastructure have been able to curb the spread more effectively by testing more broadly, Germany and South Korea being notable examples. Rather than focusing their testing only on patients with severe symptoms, they have been using their increased testing throughput to follow up on individuals that have been in contact with patients that tested positive.

Everything points to more testing being the most immediate handle that countries have available to them in order to slow the spread of the disease. Testing more people would provide crucial insights on transmission and inform strategies on how to re-open sectors of the economy in a safer way. However, how can testing facilities adapt and increase testing throughput?

We have developed a tool available at https://www.pcr-pooling.com to allow COVID-19 testing laboratories to increase their throughput of rtPCR (real-time Polymerase Chain Reaction) testing by up to 8X without requiring new equipment or training for new skills.

The basic idea is to test samples in pools that are mixed together rather than individually, and then to apply a search algorithm to more quickly identify the healthy patients in the pools. Testing samples in pools can increase the throughput of testing large populations (>1000 patients) while the infection rate in the population is low (up to ~20%).


The idea of testing pools is not new, and it has been applied in both research and diagnostic settings. However, we believe that wider adoption of pooling is limited so far by the fact that keeping track of all the pools and the order in which samples need to be mixed is a complicated process.

We are working towards making this approach more accessible to testing laboratories by providing a free and easy-to-use tool that provides optimal recipes for mixing samples into pools. As no new technology nor skills are required, it is our aim to make the adoption of our tool into existing pipelines as convenient as possible, and to increase the throughput of the number of samples possible to test with the highest reliability.

There is debate as to what constitutes a safe pool size, i.e., the number of samples that should be mixed together. One concern is that mixing samples will lead to dilution of viral material and thereby reduce detection limits to levels that make it impossible to identify the presence of the virus in a pooled sample. However, a recent controlled study performed in Israel at Technion and the Rambam Health Care Campus in the context of COVID-19 shows that detection is possible in a pool with up to 63 negative samples mixed with a single positive sample, albeit with a high number of required PCR cycles (see link to pre-publication below). Furthermore, some laboratories testing for COVID-19 in Heidelberg (Germany) and at Stanford (USA) have used this approach successfully on a smaller scale.

From the point of view of cost-effectiveness, the number of tests required is much lower when pooling samples from a population with a low infection rate and can therefore result in a significant savings in the time needed to diagnose most healthy individuals and in a reduction of the amount of required material.

We are looking to team up with research laboratories to:

  • re-validate this approach
  • improve the user experience

We believe that making this strategy accessible to more laboratories will make a difference in this global health crisis. Visit our website at https://www.pcr-pooling.com for more information, to try out the Demo, or to run a Simulation to see whether this approach would work in your laboratory setting.

Please get in touch at info@pcr-pooling.com if you are interested in teaming up or know someone who might be.

Charles and Fredrik.