H/T: High mutation rate in Homo heidelbergensis
Background and information
Of the ancient human genomes that have been sequenced so far, Homo sapiens, Homo neanderthalensis and Homo denisova, Neanderthals and Denisovans appear genetically to be closely related. As there is a transition in Europe and parts of West Asia between Homo heidelbergensis and Neanderthals, this is compatible with Denisovans being another descendant of Homo heidelbergensis that moved elsewhere (to the south-east parts of Siberia).
All human genomes that do not appear to be descended from Eurasian Homo heidelbergensis instead appear to belong to a group of genomes that are ascribed to Homo sapiens. While it is possible that they all indeed belong to a group that formed after a split from a common ancestor that they share with Homo heidelbergensis, there is also a possibility that some factor may have created an unusually high mutation rate in Homo heidelbergensis. There are, for example, chemicals that occur naturally that are known to increase mutation rates.
The H/T is that a population of Homo heidelbergensis that lived before the Neanderthal/Denisovan split, and possibly continuing slightly later than the split but mostly before it, consumed naturally occurring chemicals (such as in the food) that caused a higher rate of mutations. That led to them diverging further from other early human populations that mutated at a significantly slower rate. This, in turn, gave the genetic appearance of one sapiens branch and one branch with descendants of Homo heidelbergensis when, according to this H/T, the sapiens "branch" is really a larger phyletic group from which the Homo heidelbergensis branch is a paraphyletic group separated by a higher mutation rate.
This hypothesis treats Homo rhodesiensis as an early human that was distinct from Homo heidelbergensis. The hypothesis is based on the notion that Homo rhodesiensis was native to Africa while Homo heidelbergensis came from Eurasia. The hypothesis does permit the possible existence of a branch of Homo heidelbergensis in Africa descended from ones that had migrated there, but it does not permit that they totally replaced Homo rhodesiensis.
This hypothesis predicts that all genus Homo genomes that are sequenced will fall into either the Homo sapiens "branch" or the branch of descendants of Homo heidelbergensis (or a mix of the two). If a single genus Homo genome is sequenced and found to be "super-archaic" the way the traditional molecular clock says that genomes that diverged nearly 2 million years ago from an early migration of Homo erectus would be, the hypothesis will be falsified. This hypothesis instead predicts that such Homo erectus descendants will have genomes in the Homo sapiens category overall (they may show some point differences related to the brain in the case of old or isolated forms, but overall DNA percentage no more different from modern humans than different modern populations are from each other) due to not being descended from the more rapidly mutating Homo heidelbergensis branch.
The hypothesis also predict that studies of the diets of pre-Neanderthal Homo heidelbergensis will show a significant incidence of foods that contained toxins that increase mutation rates (and the risk of cancer) that were not eaten, or eaten to a much lesser extent, by other early human populations. The hypothesis predict that at least some of the effect should be in the form of DNA mutations either directly or by impairing DNA repair, as opposed to epigenetic causes of cancer by regulation of cell division such as that caused by growth hormones or sex hormones. It is also possible that the toxins could come from bacteria in their intestines degrading their food, which can be tested by analysis of their feces combined with analysis of their food and experiments on the combination (the bacteria may need to be recreated using genetic engineering). These foods, or the combination of food and bacteria, should either not have been eaten by/present in later descendants of Homo heidelbergensis, such as Neanderthals and Denisovans, or these descendants should display a genetic resistance to them.
In the case of genetic resistance that took the form of lower mutation rate and/or higher repair of mutations, the descendants after the end of the high mutation period should display a lower degree of genetic change over time. This can be tested by DNA analysis. Genes responsible for DNA repair and the robustness of DNA should be tested too.
As modern humans with admixture of Neanderthals and other descendants of Homo heidelbergensis may have inherited and possibly been subject to positive selection for such resistance, the mutagenic effect of the chemicals may only be apparent on people with no such genes. In other words, on people from sub-Saharan Africa. This is about the importance of studying sub-Saharan Africans for mutagenic effects of types of food that were present to a greater extent in Homo heidelbergensis diets than in contemporary archaic human populations elsewhere. It can be done with standard methods for determining what combinations of gentics and lifestyle cause risks that neither would cause on its own. Since resistance may not be total, it does not predict a total absence of such mutagenic effects of such food in other human populations. In the case of the poisons forming in the gut, tests can be done by exposing human cell cultures sampled from sub-Saharan Africans to chemicals produced by recreated Homo heidelbergensis gut flora exposed to Homo heidelbergensis diets.
As DNA codes for proteins, protein analysis should be able to detect the effects on hominins that are too old to have preserved DNA. The H/T predicts that the overall proteinome of Homo erectus fossils older than Homo heidelbergensis (800000 years or older) should be in the Homo sapiens range or at least closer to Homo sapiens than to Neanderthals and Denisovans. If such proteinomes are more different from modern humans and Neanderthals than the two are from each other, the hypothesis is falsified. These predictions apply to the neutral bulk of protein characteristics, not to adaptive point differences that are small in the overall protein chemistry.
This hypothesis predicts that the mitochondria that went into the founders of a hypermutating Homo heidelbergensis population were one or more branches of a famiy tree of mitochondrial haplogrups with non-hypermutated relatives. It therefore predicts the existence of mitochondrial haplogroups, extant or extinct, that can be placed in the modern human range but also display a small number of point mutations that they share with Neanderthal and/or Denisovan mitochondria but not with most other modern human mitochondria.
One of the predictions from this hypothesis is that while the branch of Homo heidelbergensis descendants should have many unique mutations, their descent from an earlier population that was largely genetically modern should leave them with some of the "modern" genes that would otherwise be thought to be derived modern. Not all of them, as gene flow and adaptive introgression of individual genes may play a role, but some. This is confirmed by the discovery of the human version of the FOXP2 gene in Neanderthals that went against the prediction of traditional molecular clocks that it would be unique to Homo sapiens.
Demarcation from straw men
The discovery of more "species" closely related to Neanderthals and Denisovans fall within the possible outcomes of a mutated population that later split up and evolved, and would therefore not falsify the hypothesis.