Re: [DIYbio] DIY COVID-19 PCR test

In my opinion the Zhang lab post is not great. It ignores everything that would actually make it clinically relevant (e.g. sample processing). It's completely unrealistic to give people the idea that you can just have a viral load high enough to do a SHERLOCK assay without any sample processing. And the sample processing is the most difficult part. Their intention seems to be just to get a little bit of PR.

There are many issues when dealing with the virus that need to be considered. The current testing is only around 40% accurate because the viral load from throat swabs is unreliable. Even then, you need to have a sample processing step before you do the RT-PCR. The next problem is scaling. Most laboratories are not equipped to perform thousands of tests in a day. In China, regular biology labs at universities are doing the testing and they couldn't do more than 2000 a day with 50 staff. This is because sample processing is done in a P3 biosafety setting. RT-qPCR is just not readily scaleable and there will also be massive problems when African countries have to try and test large portions of the population. The current mainstays (Roche and Abbott) can only perform around 1000 tests in an automated fashion in a shift and they are dependent on their own reagents.

When trying to do community led research it needs to be planned out. Adding a bit of plasmid to a lab bench and then seeing if you can amplify it ignores many of the aspects of the virus that make it clinically relevant. What's a relevant concentration? How does DNA plasmid compare to an RNA virus? How stable is DNA compared to RNA? How do you know how many copies you're adding to the lab bench (260/280 is horribly unreliable when determining copy number)? What are really the questions that you would like to answer and how would they make a difference?