This paper is not evidence, it is speculation based on an inability to see a path other than gain-of-function lab work. These coronaviruses are slow to evolve d

This paper is not evidence, it is speculation based on an inability to see a path other than gain-of-function lab work. These coronaviruses are slow to evolve due to their proofreading capabilities but they recombine readily. The natural path, therefore, is a wild animal (likely pangolin, civet, raccoon dog, or even human) that is simultaneously co-infected by two viruses that recombine. The problem with genome data centric analysis is that no distinction is made between RNA samples and virus isolates. Key evidence for the gain-of-function hypothesis, is one or more lab isolates that match long segments of the SARS-CoV-2 genome, including non-coding segments.