As we gain the ability to manipulate genes, will we be able to effectively hyper accelerate the evolutionary process? Are we becoming the stewards of our own genome? Are we up to it?
If you are talking about genetic manipulations where genes from one species are transferred into another then I would say that we are bypassing evolution altogether. We are designing organisms at that point.
The tried and true method of speeding up evolution is to expose a population to very strong selection where the number of parents with offspring in the next generation is quite low. This is exactly how we produced the laundry list of dog varieties in a relatively short amount of time.
The gene may be the key to the protein but the protein is the key to the phenotype.
That is only half of the recipe. The pattern of gene expression, the trascriptome, is just as important. When a protein is expressed during development, maturation, and environmental challenge also has impact on phenotype.
For example, some studies have suggested that overexpression of anti-oxidant enzymes can increase lifespans. I can't find the reference now, but I am pretty sure that this has been done in mice.
Mapping the genome was the easy part. Understanding the proteomics of gene alteration is many orders of magnitude beyond that.
Fully agree. Just figuring out which genes produce which proteins is a small step. You still have to understand how those proteins are modified by other proteins, alternately spliced mRNA products for the same protein, etc.
With natural selection that is true. I think Dogmafood was referring to artificial selection though. Where we can dictate the desired outcome.
Once you introduce direct genetic manipulation it stops being artificial selection.
With artificial selection you are blind to the genetic sequence that underpins the phenotype. This is still recognizable as evolution since this is how it works outside of human influence. It is the disconnect between mutation, function, and fitness that makes evolution a top-down process. When you are directly manipulating the genome it becomes a goal oriented process, a bottom-up process.
Are you not still in effect performing "selection" by determing which genes are manipulated though?
That depends on whether or not you are selecting based on genetic sequence or phenotype. If you are selecting based on phenotype you may very well be selecting a feature that relies on the interaction of several genes instead of just one. For example, cattle breeds were artificially selected for milk and meat production, usually one or the other for any given breed. I really doubt that milk and meat production were determined by a single gene. In fact, many breeds have distinctive features (e.g. dished faces, coloration) that have nothing to do with milk or meat production. These features hitched a ride on the features that were being selected for. For example, breeders did not set out to create a black and white cow. They were selecting for a docile breed that produced a lot of milk. The result was the holstein.
Is it really though? Look at how many species are extinct.
In evolution, every lineage is an experiment. We humans tend to focus on a single species, but the lens of evolution covers a much broader landscape of biodiversity. Nature has the advantage of running billions of interacting experiments in parallel while us humans have to focus in on serial experiments, often in limited environments.
We humans have only begun to tap into Nature's method of solving problems. One of the big movements in drug discovery right now is high throughput. This allows researchers to create random molecules and test thousands of them against the target of interest. Prior to this innovation scientists try to solve the problem with brain power. They would imagine what type of molecule would work best, make that molecule, and then see if it had the desire effect (i.e. rational design). Scientists in these fields are still trying to balance rational design (bottom up design) with evolved design (top down design), but the benefits of high throughput screening of random products can no longer be ignored.