Evolution would not (realistically) happen because bacterial leaks would have too little time to undo all the changes and re-adapt to natural-only aminoacid environments, the adaptive pressure against synthetic aminoacids would go from zero in the lab to 100 in the wild immediately.
The possibility would had to be taken account if there was a natural way to keep the bacteria in gradually decreasing concentrations of synthetic aminoacids, you would then give a chance for the reversion to take place slowly and over many generations.

Find out the gene, make a few interfering RNAs candidates, blast it into the skin cells (not as hard since its a surface tissue), choose the ones that did not knock down other similar proteins and BAM! you got it.

In the results it is shown dose dependant mortality up to 100%, the differences in harm appear between species not members of a population. Most likely this would lead to replacement of the sensitive bugs by resistant species instead of adaptation.

I think is still too early to celebrate the technique, this are mostly just computer predictions that have not yet been proven in the lab, there are a lot of things in biology that are supposed to happen based in computer simulations that simply don't in a flask. If you have a million compounds and dock them to a single active site in a protein you may get around a thousand that are predicted to work inhibiting that site, when you test them you are lucky if a couple really have some kind of significant inhibition. Here they are talking about predicting a shape using hundreds or even thousands of this interactions in a single molecule, I would expect most of this predictions to result false and even those that get it right will do it only on a few cases out of many trials.

Even if siRNA works by the same mechanism in all vertebrate cells there are many steps were a difference between species can make it a failed treatment. For example there are many RNA viruses that can encapsulate replication machinery in certain types of cells, that helps avoiding innate immunity mechanism, if this happens in target cells in humans (may not be the same in monkeys) then the encapsulation would interrupt also the siRNA binding to viral RNA making it useless. also there is a posibility of interference of the siRNA against normal human mRNA important for cell life cycle hence toxicity and so on. Even the lipid nanoparticles could work less efficiently in the human body (sequestration by non target cells, lack of proper integration with cell membranes, activation of unusual immune responses not seen in 20 monkeys but maybe fatal in a fair percentage of humans, etc.)

Yes, it is most likely that the treatment can be useful according to their results until now, but there are literally hundreds of reasons why this could be not the case, it would not be the first nor the last time when a perfect treatment for monkeys proved to be useless in humans.

Its not so easy, viral vectors work well to deliver nucleic acid sequences to act directly or by the proteins that they encode.

This approach unfortunately depends of the combination of a protein and a sequence of RNA, even if you can make a viral vector that encodes both the RNA sequence and the nuclease so they can be produced, there is no process that can be used to combine them both inside the cell, so they cannot function.

A good delivering method effective for this kind of approach would also allow several other very effective cure candidates that have exactly the same problem. It is an interesting technique but not exactly an advance.

Measles virus is the one that apparently puts the cancer into remission. If the MMR vaccine stops the virus, then the vaccine "causes" cancer.

It's a valid question, when looking to the simplest organisms such a viruses or bacteria you can observe repetition of specific adaptations when under the same environmental pressures (up to a certain point). You neutralize several strains with monoclonal antibodies or a compound against a specific protein and the escape mutants frequently show the same changes. The question is "how much of this is conserved in more complicated organisms?" It would not be the first time that an apparently chaotic process was actually following some hidden rules.

So yes, it seems that adaptation is quite chaotic in macroscopic life, but since there was a possibility that this was not the case it has value to confirm it.

At least 100 more resources are being used for dengue than for mosquitoes, unfortunately for dengue having "near perfect" protection (the normal situation for all other vaccines) is not only not effective, it actually produces a worse disease. For better or worse controlling dengue is going to take a few more billions and at least one more decade. Also, you control the mosquito and you control several diseases at the same time.

The problem in this case is not so much the danger of the genetic manipulation (the approach seem to be based in minimizing risk) but seen how effective it is really going to be in a large scale situation. People worry much more about this being a waste of money than a danger to the ecology.

Also, the process specifically make the females produced by this males to become sterile so for one part you will get slowly more and more gene-carrier males competing for the healthy females (that will be less and less frequent) in every generation, it will have the extra merit of making the affected females less prone to bite so the risk to humans dimish.

Anyway, the good thing is that this approach affects only a single species of mosquito so even if this goes out of control you have very few risk to the ecology, compared with other much more risky trials (like those done with the Wolbachia parasite in Australia) this seems to be relatively safe.

Visit an ER or an ambulance with paramedics for half a day, you would be surprised of how much can be done for people even when you have no time or access to equipment and most drugs.

As long as the cells produce a large amount of proteins any plant could be used, but Tobacco has been studied extensively so its easier to modify, also Tobacco already have a negative image so it will not change with the GM. (never mind also that every plant produce potential drugs). Nevertheless the research strikes me as too complicated, they put a lot of effort to explain how their method minimize Antibody dependent enhancement (ADE) but that has never been a problem for WNV or related flaviviruses, only for dengue (and for a reason that is completely absent for WNV). Also, there are several vaccine candidates for humans in trials that are produced by traditional methods much easier to prove safe (inactivated virus vaccines seem to be effective enough). So by the time this GM tobacco antibodies are ready for use most likely the population in risk will be already safe and the only people in need would be those without normal immunity.

As in the case of Yusuke Katayama, Japanese law enforcement proved to be quite ignorant about technology crimes. After getting death threats on messages boards they managed to "get" confessions from several people that later were proved to be just victims of malware in their computers. It is normal to have doubts about their capacity to deal with cybercrime.