great ape: Could you perhaps elaborate on how this statement doesn't contradict itself?

All the basal processes originated early in evolution, but still in a stepwise fashion. Replication is essential and should be one of the first things that arose. I am more thinking in the line of 'frozen accidents' than a slow gradual evolution. In any case not a special Creation event.

Could you perhaps elaborate on how this statement doesn't contradict itself?

chunkdz: I am skeptical of the explanatory power of molecular genetics in the light of evolution. I am even more skeptical of the predictive power of molecular genetics in the light of evolution. How are either of your above statements distinguishable from gross speculation?

I do think that replication and cell growth have evolved, but needed to have evolved very early in evolution. Replication should have been there from amost the beginning, just as fault-tolerance and decisions about cell divisions. We have to find the molecular scenarios that fit with the origin of these systems.

I also think that the interactions between developmental systems can predict the path of evolution, and is therefore more than just speculation. To specify this, I model evolution on a design pattern called design-by-contract. The sequence of developmental pathways predicts the way the sytem was assembled in evolution. That's the connection between development and evolution: ontogeny follows phylogeny in molecular terms.

The developmental systems, like replication and cell growth did not arise suddenly, but they evolved gradually…

This seems counter to what you write later:

…control of cell division has to have originated at the origin of cellular life itself.

I am skeptical of the explanatory power of molecular genetics in the light of evolution. I am even more skeptical of the predictive power of molecular genetics in the light of evolution. How are either of your above statements distinguishable from gross speculation?

Having knowledge of mechanisms is possible if you know nothing about their historic causal roots. We do know quite a bit about cellular division. What we are looking for are different causal pathways that lead to cancer. Understanding how something was put together is no guarantee that you can figure out what is the cause of something going wrong. It helps but they are distinctly different conceptual problems.

I think it is safe to say that development and evolution are closely linked. The developmental systems, like replication and cell growth did not arise suddenly, but they evolved gradually and every step should be functional (the principle of functional continuity). Only seen in the light of evolution, the system can be unraveled, because only then we can see why things have their place in the system.

Because cancer is so much connected to normal cell division, control of cell division has to have originated at the origin of cellular life itself. The ultimate control mechanism of replication thus seem te reside in ancient pathways. In order to effectively interfere with cancer, you have to have knowledge of these ancient mechanisms. And since we don't know how replication and early cellular signals evolved, we can safely say that we lack the elementary principles of cell division and thus the insight to cure cancer at its roots.

An important focus in cancer research is to try to get the cells enter a genetic differentiation program. This important switch (stem cells vs differentiated cells) has also been an important step in the evolution of multicellularity. Here as well, unraveling the fundamental steps in evolution will lead to the understanding of this system and targeted interference with cancer.

In order to unravel the events that cause cancer, we need to know how the different subsystems interact and we can only find out if we know how they were assembled during evolution.

Why?

bfast: I would say, rather, that neo-Darwinian evolution can only be validated, or invalidated, by molecular genetics.

And the funny thing is: it has been invalidated because the only semi-molecular explanations that I know of suffer from the problem of irreducible complexity. Molecular evolution (genetics) is about how molecular systems and subsystems evolve and communicate with each other.

Neodarwinian scenarios do not translate into molecular sequences of events. Therefore, molecular genetics invalidate molecular evolution.

bfast: There is much to learn about molecular genetics that has nothing at all to do with evolution.

If you mean that we have learned a lot about development despite the fact that we don't understand evolution, OK. However, if you don't take evolution into account, we will never truly understand development. Part of this follows from the teleological aspects of evolution that are reflected in developmental processes. It's the discussion between descriptive and predictive science.

For instance, we know about the checkpoints for cell division in eukaryotic cells (descriptive). We do not know how the eukaryotic cell originated and why the different check points were put there in the first place. In order to unravel the events that cause cancer, we need to know how the different subsystems interact and we can only find out if we know how they were assembled during evolution. The failure to understand the evolutionary processes will be reflected in our understanding of developmental processes, like cancer.

This is easily seen in the publications in Nature, Science and Cell. With every new molecular process discovered, the message is always that it is even more complex than we imagined. With a good predictive theory, however, this should be the other way around. With new discoveries, we should be able to link pieces of the puzzle and move forward. Predictive theories must be made on molecular pathways. Life is molecular.

Bradford: In any case any evolutionary events that do not attempt to describe underlying molecular events are all the poorer for it in my view.

This was exactly my case. Evolutionary explanations, especially for the macroevolutionary events are not done in molecular terms and are therefore speculative, if not useless.

I would say, rather, that neo-Darwinian evolution can only be validated, or invalidated, by molecular genetics.