Just became aware of an article from May interviewing experts to debunk lab scenarios for SARS-CoV-2 origins (it was cited in an Atlantic article published today). I'm going to work through the arguments here to see if any new data has shifted the balance. https://www.cidrap.umn.edu/news-perspective/2020/05/scientists-exactly-zero-evidence-covid-19-came-lab

The article reviews 2 conspiracy theories: (1) bioweapon, (2) release of natural virus being studied in the lab. I'm going to address points raised for both conspiracies, but it's not possible to know if a virus was designed as a bioweapon unless you have proof of intent.

So the first overarching claim is that there is "exactly zero" evidence that COVID-19 came from a lab. I agree. There is also exactly zero evidence that it came from an animal or was pre-circulating in humans before emergence in Wuhan in late 2019.

The next argument is that several SARS2-like viruses have been found in bats and pangolins and only recently published in late 2019 and 2020. Therefore, there are SARS2-like viruses in nature.

Let's look at the SARS2-like viruses. The first to be discovered, RaTG13, came from the WIV's search in 2012-2013 of one cave in Yunnan for a pathogen that killed 3 of 6 miners who contracted a SARS-like viral infection back in 2012. WIV sequenced parts of RaTG13 in 2017 & 2018.

Another prominent SARS2-like virus is the Guangdong pangolin CoV which was characterized by 5+ papers. Mainly, a single sample was the source of the majority of the CoV genome seq. There is another SARS2-like pangolin CoV from Guangxi, but doesn't share similar RBD to SARS2.

In any case, both CoVs were from smuggled pangolins, unclear whether they got it from the smugglers (humans) or other animals in captivity. Hope the WHO can track down at least the Guangdong smugglers to find out - the pangolins were confiscated in March, 2019, not too long ago.

Two other less closely related viruses, ZXC21 and ZC45, were described in this paper, late 2018, wherein bat intestines were ground up and injected into the brains of live suckling rats to isolate and characterize the natural virus - replication-competent. https://www.ncbi.nlm.nih.gov/pmc/articles/PMC6135831/

One other virus, RmYN02, was assembled from a metagenome of a pooled sample (11 fecal samples collected from bats in summer of 2019, again in Yunnan). Parts of its genome were sequenced from a single sample known to carry multiple closely-related CoVs. https://www.cell.com/current-biology/pdf/S0960-9822(20)30662-X.pdf

What all of this tells us is that the most closely related viruses to SARS2 can be found in Yunnan. But it is not evidence of the route by which SARS2 emerged in Wuhan, more than 1000 km away where 0% of the population had SARS-reactive antibodies (2015). https://www.ncbi.nlm.nih.gov/pmc/articles/PMC6178078/

In terms of circumstantial evidence, it's clear that these SARS2-like viruses were both found in nature and under active study in multiple labs across China. It doesn't tell us anything about SARS2 itself.

3rd argument in the CIDRAP article: scientists don't know enough about CoVs to engineer de novo function/proteins. For SARS2, this can be broken down into (a) The SARS2 RBD was found in one 2019 pangolin CoV. (b) Our computer model couldn't design this RBD.

Having already discussed what the discovery of SARS2-like viruses means, it leaves us with 3b. I agree. Our models don't have enough data to learn from to design this RBD de novo. I don't think the SARS2 RBD is an original human invention.

This leads to the next point 3c. There are no known (published) virus backbones to base SARS2 engineering on. I also agree. It's also impossible for models to detect whether viruses have been engineered from naturally existing viruses that we don't know about.