Long story short, it involves mutations in viral proteins that are responsible for counter-acting anti-viral proteins (termed restriction factors) in human cells, that differ slightly from the chimpanzee versions due to an ongoing genetic arms race between the host proteins that block viral infections, and the viral factors that counter-act them to retain virus infectivity. There are also changes in the viral envelope proteins to help evade adaptive immune responses (recognition by immune cells and antibodies)

That would be easy enough to test, provided we had archeological samples of Phoenicians / Romans who were infected with M. tuberculosis (Mtb). Then, one could do the same phylogenetic analysis done in the paper that claded with the seal / sea lion Mtb sequences. Of course, the sequence analysis data provided in the paper would probably argue against your hypothesis, as Mtb sequences obtained from infected members of those civilizations would probably clade with modern European Mtb sequences rather than with seal / sea lion Mtb, which it would have to do in order to fit the data in this study.

Wrong.

Step 1: Use public funds to do innovative research into expanding the genetic code in microbes.

Step 2: Patent everything to make sure no one else can build on your discoveries.

Step 3: Create a company that promises all the keywords for a biotech e.g. antibiotics, vaccines, etc.

Step 4: Get bought out

Step 5: ? Profit was Step 4.

Remember when science was about discovery and standing on the shoulders of giants?

Monsanto is, in part, doing exactly what you suggested. The genetic element that grants their crops resistance to glyphosate (Round-Up) was discovered in microbes growing in waste runoff containing glyphosate. The patent is on the plants that have been transformed / engineered to contain this naturally occurring resistance gene in their genome and express it to garner resistance to the herbicide.

The cDNA argument is much worse to ludicrous. The only thing one could imagine is patentable surrounding the issue of cDNA is the technique involved in its generation, but that ship has sailed long ago. The entire process of generating cDNA is by using materials all found in nature. The RNA template that is used to generate cDNA in these cases of naturally occurring genes is obviously naturally occurring* and the technology to even create cDNA in the first place is using a naturally occurring enzyme, reverse transcriptase (found in retroviruses like HIV to catalyze the conversion of their genomic RNA to DNA). Because all these elements exist in the natural world, it is certainly possible in say HIV infected humans that random gene mRNA molecules have been converted to cDNA, thus negating the argument of generating something that does not exist in the natural world. More importantly, the only facet of this whole process of generating cDNA that is artificial is placing the materials in a tube together, NOT inventing any novel chemicals or enzymes to catalyze the process!

* In cells, mRNA molecules are heavily modified, sections spliced out, nucleotides edited, the 5' end capped, the 3' end poly-adenylated, so with the arguments placed forth concerning cDNA, one can imagine in vitro transcription of a gene creates an RNA molecule that doesn't exist in nature and can therefore be patented. Ridiculous, right?

As far as I know, bacteria don't use epigenetics in the same way we would associate with eukaryotes. First off, bacteria don't have histones and are thus limited to DNA methylation in terms of types of epigenetic modifications they can employ (bacteria do have DNA-binding proteins that help package DNA, but I don't believe anyone has shown they are used to control gene expression in a heritable manner -- the definition of epigenetics). In terms of how bacteria use DNA methylation, there are two cases I am aware of: (1) distinguishing between newly synthesized and "template" DNA just after DNA replication to help fix errors (if the DNA repair machinary detects a mis-matching base-pair, it corrects based on the template, methylated strand; and (2) to distinguish between self and non-self DNA, particularly in anti-phage (viral) defense, so that enzymes that have evolved to destroy foreign DNA can avoid destroying the bacterial DNA.

In general, epigenetic markers are controlled by protein interactions, which can of course be very complex, including kinetic competitions between proteins that write the markers, and proteins that erase the markers. To build a genetic pathway that regulates epigenetic markers beyond simple sequence recognition (such as GAmTC methylation in E. coli that regulates the origin of DNA replication) is probably beyond the capability of current synthetic biology.

I'm not sure about influenza (BSL-3 requires a respirator, and since influenza is very easy to catch through aerosol, makes sense to protect against breathing it in), but HIV at the University I work at is considered BSL-2* (that's BSL-2 star). Basically BSL-3 minus the breathing apparatus, as you cannot contract HIV by breathing it in. So you double glove, no sharps, and bleach everything down, and make sure everything in there stays in there unless you are absolutely sure the virus is dead.

Mice aren't humans, human experimentation is still morally objectionable and illegal, and medical testing on primates / apes is much more expensive and considerably less ethical in most people's minds. It's important to note where mouse research fails to properly recapitulate human biology, but sensationalist journalism acting as though the animal rights crowd is finally vindicated in proving, through a few heroes in the scientific community, that animal testing is cruelty without merit is harmful to the field, both in terms of basic biological research and discovering new medical treatments. Mice are still the only mammalian model organism where gene knockouts and knock-ins are reliably possible (some new DNA modifying technology involving proteins called zinc fingers are changing this) and this study is only demonstrating that some aspects of mouse research cannot be translated into human medicine. And for those questioning how many drugs failed in mouse models that may work in human subjects, are you going to be the group / company testing drugs on humans that are toxic or otherwise not working in mice? Would you rather kill an equivalent number of chimpanzees as we do mice to get a more accurate model of human biology, ignoring the added cost of housing and longer time of maturity?

I'd only add:

Observation -> Hypothesis -> Theory -> Predictions -> Test -> Scientific fact.

Well, yes, the headline is misleading, but it's also a bit more than a "possible" ancestor.

The researchers in the study wanted to create a better phylogenetic reconstruction of the evolution of mammals than had been previously accomplished, to resolve whether divergence of placental mammals from non-plancental mammals (egg-laying / marsupials) occurred before or after the extinction of the Dinosaurs (the K-T boundary), and also to make predictions of the biology of that last common ancestor. Previous phylogenetic reconstructions had been done with molecular data (DNA or protein sequences), but molecular data is limited to extant species and makes a lot of assumptions about the rates of changes in DNA that get more unreliable the further back in time you go. This study combined molecular data with character traits they call 'phenomic' characters - from the paper: "4541 phenomic characters de novo for 86 fossil and living species." The resulting matrix of traits, both molecular and character, was used to generate a tree based on maximum parsimony - a method which minimizes the number of trait changes over time when building a tree. This resulted in a single, highest scoring tree predicting the evolution of these species and the changes in their traits over time. The resulting tree is then "clocked" (called 'time-calibration in the paper) to known rates of evolution for the molecular data (good for recent divergence of species) and by fossil data to give time ranges for the deeper sections of the tree. This last part is key, as you cannot get molecular data from fossils, and fossils allow you to map the existence of certain traits within a group to a certain point in the history of these organisms.

The result is a time-range in which the last common ancestor between placental and NON-placental mammals must have lived, given the data provided and the parsimony criterion. As the tree makes claims about when the phenomic characters evolved or were lost, it also predicts which phenomic characters the last common ancestor had.

More specifically, the chemical is a modified version of adenosine called cyclic AMP (cAMP). cAMP is generated by enzymes from ATP, the energy molecule of cells. It acts as a signal amplifier in cellular signaling cascades, and for the purposes of how caffeine affects the body, increased cAMP concentrations = more cellular metabolism. Think the adrenaline response at the cellular level.

Caffeine inhibits enzymes which break down cAMP and turn off the signal. No off switch = artificial build-up of cAMP in the cell = artificial high metabolism state.

Cocaine acts in a somewhat analogous manner, in that it doesn't increase the amount of neurotransmitters being released directly or directly stimulate neurons, but prevents released dopamine from being broken down and thus prolongs the signals artificially, leading to the psychological effects of cocaine use.

At first, I was super excited by the headline and thought: "I hope they include these newly discovered python viruses!" Only to quickly realize the authors meant a different kind of Python...

"[using] a disabled form of the virus that causes AIDS"

While true, this is a poor way to describe a lentiviral vector, meant to invoke the idea of using HIV to kill cancer in the minds of readers not familiar with modern molecular biology. HIV is a type of virus called a lentivirus, which itself is a type of retrovirus, which means that it takes the RNA genetic code it has packaged in the virion, chemically transforms it into DNA, and integrates this DNA into the DNA of the infected cell. Lentiviral vectors are designed such that they do this part of the viral life cycle, but are engineered to lack the genes necessary to make more viruses, so the integrated virus is dead on arrival.

In this case, the researchers kept the normal HIV surface receptors so the virus would efficiently target and "infect" T-cells from the patient; normally, lentiviruses are given a generic non-HIV receptor so they can infect any cell type you might be using in your lab experiments. The lentivirus genome contained not the normal viral genes, but a chimeric T-cell receptor designed to stimulate an immune response against CD-19, a surface protein specific to B-cells. Once this chimeric gene is integrated, the T-cells will express it on their cell surface, and stimulate the immune system to target and destroy cells that have CD-19 on them; this kills all the B-cells in the body, both healthy and cancerous. This last point is a problem brought up by TFA, that the patient now essentially has a limited auto-immune disorder as the altered T-cells persist in her body and continue to point them immune system to targeting B-cells, leaving her partially immuno-compromised (which is the funny part about using the "virus that causes AIDS" to do this).