I found a new paper (it is not in fact new, it was published March 2020) from ResearchGate. Also this paper explains the evolution of SARS-CoV-2 by recombination
S.U. Rehman et al, “Evolutionary Trajectory for the Emergence of Novel Coronavirus SARS-CoV-2,” Pathogeny, March 2020.
My interest to this topic is not so much Covid-19 but evolution in general. Mathematically it is fairly easy to show (as I have done in some posts that demonstrate the problem in evolution. As I consider darwinism a religion, they are in the category religion) that mutations alone cannot explain how a new protein coding gene emerges. Evolution of new species by point mutations to control parts in genes can turn on and off existing genes (existing in junk DNA) and in this way fast changes are possible, but some steps in evolution require new proteins and therefore new protein coding parts of a gene. In the SARS-CoV-2 virus such a case is the S gene which codes the spike protein. In order for this virus to infect a new host (like to jump to humans) there is needed a significant change to the S gene. This significant change cannot be easily explained by a sequence of mutations (it is too improbable to get a working sequence), so all papers that have studied the evolution of SARS-CoV-2 have suggested recombination events.
SARS-CoV-2 is a RNA virus and RNA is more prone to mutations and recombination than DNA, but evolution of host species can only happen with changes in the DNA. Therefore it cannot be made with RNA viruses, but it could be made with DNA viruses. There are even viruses with both RNA and DNA parts, where the fast mutating RNA part may offer an easier way to infect the host. At least some DNA viruses, like the Human Papillomavirus (HPV) can occasionally be merged to the DNA of the host. It follows that at the subsequent breaking of the cell where a virus has merged with the host DNA there may be a recombination event with another virus DNA, which takes a small part of the genome of the host to a new recombined DNA virus. This virus can infect a host of a different species, possibly much later, when the virus has gained access to a new host. By merging its DNA to the host in a rare event, it can transfer a gene from one animal species to another. This is the only way I understand that it is possible (as has been found out) that humans share more genes with domesticated (like cows) or cohabiting animals (like rats and mice) than with animals that do not have contact with humans, though the genetic distance from a phylogenetic tree is the same in both cases.
I think a sounder theory of evolution could be created by studying the possibility that infections of DNA viruses would have a larger role in fast evolution of new features at some points in the world prehistory. This is a good topic for a dissertation maker, I think. What is always missing from the treatment of genetics is calculations that show that the type of evolution they suggest is mathematically possible, i.e., not too improbable. This is also the case with the mentioned paper of SARS-CoV-2 evolution. It should be possible to make a mathematical model and to show that the evolution scenario can happen in finite time.