Testing Selection Models: Part One

Wedge with iDesign@UCI has posted a message entilted 'On Evolutionary Explanations.' I've included the last three paragraphs for the purpose of adding commentary. Wedge's remarks are italicized. Mine are in standard form.


Take this model of the evolutionary origin of the bacterial flagellum. It is long, detailed, and testable. But just what aspects of the model are testable? Precisely the aspects related to homology. The testable claims are about which components may be more ancient that others, which structures may have been intermediate between the Type-III secretory system and the bacterial flagellum, etc. The question of whether natural selection acting on random variation is powerful enough to make the hypothetical transitions is never addressed. Instead it is assumed that proposing intermediates and telling a story about how each might provide an incremental fitness benefit is enough.

The referenced article attempts to identify homologous proteins, suggest mutations that would enable cooption and identify the basis for selection. Testing of protein structures is of course possible. Identification of different types of mutations is also not difficult. The last endeavor makes or breaks the plausibility of the model. The identification of homologous proteins is consistent with the notion that proteins, believed to have evolved, had precursor functions. Contemplated mutations could be viewed as those that, for example, would produce new binding sites. Such binding sites would enable protein interaction which at first might not be optimal but which would gradually become more efficient through selected mutations.

A pattern is in evidence at this point which will be observed repeatedly. Genes encoding proteins with a specified function undergo mutations leading to the encoding of modified proteins that acquire new functions. This is a thought pattern. It is important to grasp the implications of this before proceeding further. The concept that mutations of existing genes lead to new protein roles, through a gradual modification process conferring increasing functional utility, is a logical argument. It must yield to testable hypotheses before the argument acquires the needed empirical basis. The type of testing required to document the theorized role of natural selection in a cooption process is not satisfied by suggestions from the article.

The author suggests evaluating the role of natural selection through studies of analogies and mathematical modeling. The difficulty with analogies is they tend to be predicated on the same homology circumstances as the model in question. An assumption of their plausibility is built into the assessment process. Mathematical models can be helpful but their reliability is input dependent. The data provided must encompass all relevant variables. A missing one can signify an erroneous outcome. In the end there is no substitution for submitting a biological system to testing.

The type of tests needed are those that show the predicted outcomes. An evolutionary theoretical pathway to irreducibly complex systems argues the viabilty of a model showing gradual adaptative improvements such as was indicated by the binding site description. There should be stages along a pathway to the flagellum structure identified with specific intermediate functions and the protein complexes that enable them. There is a need for tests that demonstrate functional intermediates with biological utility consistent with an incremental build up model. No small amount of bio-engineering is required but the effort is hardly superfluous. Natural selection has been a logical construct since introduced by Darwin. It needs to be placed on a sound empirical footing if concepts like precursor systems and gradualism are to be taken seriously. It is one thing to contend that small incremental changes result in new functions through selective intermediates. It is another matter to demonstrate this experimentally.

A more difficult test would entail simulating an actual mutation process. Mutation rates and mutation types yield to statistical analysis but this looks challenging given long time frames. Nevertheless as the article states: "Science is advanced by proposing and testing hypotheses, not by declaring questions unsolvable." Scientific advances are linked to the production and utilization of biotechnology as well.

The author concludes that proposing an incremental Darwinian model, he calls plausible and testable, undercuts the idea that "extraordinary explanations" i.e. an intelligent inference is plausible. The plausibility of the model is outcome dependent. The outcome of what tests though? Tests of the capacity of natural selection to generate systems through cooption. Anything less makes an end run around Behe's real point; the empirical sufficiency of Darwinian models.