Post has attachment
#birdtracheaswhat are you doing like a goddam French horn in that bird

You would think it has to be exactly like a French horn acoustically, tuning the resonant frequencies of the tube. But the articles as summarized make only one mention of lower formants -- mostly they seem to be looking for louder sounds, which I don't expect as a first-order effect.

We need some spectra around here.

Post has attachment
Look at how this budgerigar trachea just sort of loiters in the neck, to the right-hand side of the spinal column. Or the other side, whatever.

Post has attachment
#todayilearned  that in the Leach's storm petrel, older birds have longer telomeres than younger birds. This is unusual.

But do an individual's telomeres actually lengthen with age? Or do they not lengthen, but the individuals who started with shorter telomeres selectively die?

A study on hatchlings and old birds finds some evidence for the latter 'Selection' hypothesis: hatchlings have a wider distribution of lengths, and length does attain a similar maximum as in old birds. They also find something else unusual, sampling the telomere population of each individual: the old birds don't show any accumulation of short telomeres.

The authors' unifying explanation is that this species shows approximately no telomere shortening with cell division. This allows us to see the Selection hypothesis showing up clearly.

The elephant I think this leaves in the room, here: by what means do short telomeres cause lower survival? The 'Hayflick clock' hypothesis was that telomeres shorten with each cell division until they're all gone. That picture has been complicated, but doesn't this paper suggest it's more radically lacking? If a hatchling bird with medium-length telomeres is going to maintain constant length and then tend to die around middle age.

It would be helpful (though tough with a pelagic species that nests in the back of beyond) to have longitudinal samples of some birds. Through life, ideally, to see year-to-year survival and telomere length. (with PDF)

Post has attachment
Oh. My. God.
It's true.
Look at it.

Post has attachment
TLR13 binds the RNA sequence "CGGAAAGACC". What's that? That's in the ribosome of many, many bacteria, and not in the eukaryotic ribosome. It also is the target of erythromycin, which gloms onto there and blocks the ribosome from making proteins. Erythromycin-resistant strains also avoid TLR13 detection. Other than that it's a great trick, excellent precision/recall curve.

TLRs are found in multicellular eukaryotes, but erythromycin is a weapon of bacteria against bacteria. Independent developments, and the actual binding tip of TLR13 is totally unlike erythromycin in structure. It's just multicellulars' bad luck that bacteria had long ago done some spoiling of this trick we found.

Post has attachment
AMPs, anti-microbial peptides, a wide variety of short proteins that kill bacteria (and other unwanteds). Basically it's like we make our own antibiotics, yes, but these tend towards less finely-tuned to a target, more brute force. They still are good enough at hitting specific targets -- they're druglike enough -- that it makes me wonder just how much of pharmaceutical design could theoretically be done by evolving these things. Do they have enough range in chemical space?

The antibiotic gramicidin D, as found in polysporin ointment, is an AMP produced by a soil bacterium. The reason that it's an external ointment is that it'll kill our cells too. AMPs that animals produce are presumably likely to be more safe for animal cells, but I wonder if we're toxicity-limited on our own use of them.

The classical AMPs, and this covers a good chunk of them all, seek bacterial cell walls because the AMP is positively charged and the cell wall is negatively. Then they destabilize the membrane or punch holes into it.

It turns out not all AMPs are 'classical' antimembrane agents. Some enter the bacterium and inhibit its DNA synthesis by binding to something. Some inhibit RNA synthesis, or protein synthesis, or cell wall synthesis, or specific important enzymes.

(AMPs are oligomers, like 10 - 100 amino acids, and they act more like drugs than like enzymes. Lysozyme is an antibacterial protein, but it's an enzyme (the original enzyme understood), chomping up cell-wall polymer.)

Random articles:

Post has attachment
Interesting line of argument that amyloid might be a weapon of the innate immune system, finding that bacteria in the brain appear to trigger deposition and seed plaques, and on the other hand amyloid is effective to limit infection.

The NYT article didn't point this out, but the "seeding plaques" work was in 5xFAD transgenic mice, which develop amyloid pathology by a few months old regardless. You have to start somewhere, but what are the odds we're seeing an artifact of that? (Have previous discoveries in 5xFAD mice translated well to the real disease?)

Post has attachment
Sorry, maybe that cool stuff about toxoplasmosis -- Jedi mind control parasite leads to association with particular personality traits, cognitive effects, and schizophrenia -- doesn't pan out.

Pretty good-looking study, longitudinal on a cohort of 1000 people followed from birth: that's about as quality as you can get without randomly infecting people. Many studies that showed effects have been based on a clinical sample (patients with schizophrenia, e.g.), where you find they have elevated toxoplasmosis rates relative to a more-or-less-matched sample of the general population. For some reason there were Buzzfeed stories about those finding an effect, but not about this not finding an effect.

Post has attachment
Does anyone know, is the human immune system's degree of sensitivity to lipopolysaccharide a mistake, or does it make sense? It has always sounded like the immune response kills more people than Gram-negative bacteria ever would. And if we posit that immune systems are rational, what is it about the environments of mice and people that makes them 1000x less sensitive to LPS?

(Background: LPS is a building block of bacterial cell walls, so practically any living thing with an immune system targets it, including mustard plants.
But the response varies by species, and humans have dialed it up so that a microgram exposure will send us into septic shock. When we're infected by these bacteria, the symptoms are usually not from their direct effects, but our immune system firebombing the whole block to get them. Speculatively, LPS-triggered inflammation might cause chronic effects, too, like insulin resistance, and MS flare-ups.)

Is this response in fact wise? Were our ancestors subject to such aggressive attack by Gram-negative bacteria that it was wise? Or is this a bug, maybe a side effect of tuning the immune cascade for something else?

Post has attachment
We have a protein, TLR9, that detects pathogen DNA by looking for a very simple thing: C followed by G. The human genome is about 21% C and 21% G, so you'd expect about 4% C-G, but the actual frequency is 1%, all vertebrates have this C-G deficit. Bacteria and viruses have 'normal' frequencies, so looking for this is a good heuristic for the immune system.

C-G is found in "islands" of high frequency, frequently in gene promoters. The rarity is in coding material.

Wikipedia suggests the reason that vertebrate DNA is low in C-G is that vertebrates do C-methylation, which gives the C a risk of spontaneously changing into a T. Several things here I don't follow about that, though:
1) Why is this an issue for C-G specifically, instead of for all C?
2) And why an issue for vertebrates specifically, since bacteria do C-methylation too?
Wait while more posts are being loaded