Cavalcade of Mammals

I want a pet book worm!

The North Auckland worm, Spenceriella gigantea, reportedly glows in the dark brightly enough to read by!

In 2005, a Schweitzer et al. reported that they had found soft tissue in the marrow of a T. rex. bone. Examination of the tissue showed evidence of cell-like structures that could indicate preserved cells. Also stuff was squishy like protein. It if held up, an incredible discovery.

In 2007, Schweitzer et al. followed it up with more extensive analysis that indicated the presence of collagen I protein. Most exciting, they used mass spec to sequence a few fragments of dinosaur protein. It was quite similar to chicken collagen, and of course birds are the closest living relatives of dinosaurs.

Then in 2008, Kaye et al. analyzed a number of fossils and found similar traces, but interpreted them as bacterial colonies, as biofilm. They even saw the cell-like traces in a fossil ammonite, a creature without blood cells.

This year, Schweitzer et al. doubled down, publishing collagen I protein sequence from a hadrosaur.

In total the two papers found protein sequences that cover 169 aa of collagen protein sequence. Here are the overlapping dinosaur sequences with the gaps removed aligned to chicken sequence. The dinosaur sequences differ from the chicken sequence at five positions.

It will be interesting to see if these results hold up. I would like to see replication by a second lab. Most convincing would be two labs given the same unknown sample and both finding the same new collagen protein sequence.


>Tyrannosaurus rex
>Gallus gallus
>Brachylophosaurus
GATGAPGIAGAPGFPGARGAPGPQGPSGAPGPKGVQGPPGPQGP--------------------------
GATGAPGIAGAPGFPGARGPSGPQGPSGAPGPKGVQGPPGPQGPRGLTGPIGPPGPAGAPGDKGEAGPPG
GATGAPGIAGAPGFPGARGPSGPQGPSGAPGPKGVQGPPGPQGPRGLTGPIGPPGPAGAPGDKGEAGPSG
                   **                                               * 

--------RGSAGPPGATGFPGAAGRGVVGLPGQR-----------------------------------
PAGPTGARRGSAGPPGATGFPGAAGRGVVGLPGQRGETGPAGPAGPPGPAGARGSNGEPGSAGPPGPAGL
PPGPTGAR-GSAGPPGATGFPGAAGR---------GETGPAGPAGPPGPAGARGSNGEPGSAGPPGPAGL
 *

-GLPGESGAVGPAGPPGSR
RGLPGESGAVGPAGPIGSR
RGLPGESGAVGPAGPPGSR
               *  

I finally read the iconic dinosaur book by Michael Crichton. The book was very similar to the movie, closer than any other movie I’ve seen. In many ways the book reads as if it was written with the idea of turning it into a movie in mind. The plot is straightforward: rich old guy hires scientists to recreate dinosaurs from DNA preserved in fossils, then the dinosaurs get loose and eat people.

The science fiction idea than spawned the book is grand. Recreating dinosaurs! Real dinosaurs! That people can be see and watch and eventually run screaming from. The other part of the book, the horror movie bolt on plot, is naturally fit for a movie.

Surprisingly there isn’t much more to the book than what’s in the movie. And unfortunately the worst parts of the movie are the author’s invention. The ‘mathematician’ character, spouting ridiculous idea that chaos theory proves everything will go wrong and fall apart is all the author’s. Also, the annoying younger sister who alternates between fear, whining, and suicidal stupidity is all Crichton. She’s written worse in the book, the other characters mock whatever she has to say and keep telling her to shut up. The out of nowhere scene in the movie where she pops up as a computer system expert looks added in an attempt to give her character a positive side.

Still, dinosaurs!

When the book was written, it was plausible to speculate that fossils millions of years old would contain bits of DNA. As it turns out, DNA degrades over hundreds of thousands of years, and no DNA has been recovered from samples millions of years old. In fact, chemical studies predict that DNA will degrade at such a rate that no original DNA remains in samples millions of years old. Today, alas, it seems unlikely that dinosaur DNA sequences will ever be recovered.

Two terabyte hard drives were recently released, which to me means the day is nearing where a single computer with have all the storage space I can imagine ever using. This figures a typical computer with four hard drives. Here’s how I apportion a my complete storage needs:

1 TB, 100,000 books x 10 MB each, every book of which I’ve ever heard (or read).
1 TB, one million photos x 1 MB each, a lifetime of pictures and LOLcats.
1 TB, 200,000 songs x 5 MB each, every song I’ll every hear in my life.
7 TB, 200 TV shows x 50 episodes x 700 MB, complete runs for typical hour long TV shows.
12 TB, 3,000 movies x 4 GB each, one new movie a week for life.
——————————————————————————————-
Total: 20 TB

Four 5 TB drives add up to 20 TB, and 5 TB drives should be available in about 2-3 years. Backup and redundancy require extra storage not accounted for here, but becoming progressively easier.

As you can see, books and even audio are only a small fraction of the storage space. Start with a smaller video library–movies and TV seen so far, or only favorites–and four 2 TB drives would satisfy all personal storage requirements.

To use more personal storage someone needs to make some sort of digital diary, an elog, a continuous lifetime audio or video record.

Putting together the reading list for the seminar class this semester I ran across an odd corner of the human fertility industry that I never knew existed. I’m picking seminal papers in biotechnology and one of them is the 1992 paper, “Birth of a normal girl after in vitro fertilization and preimplantation diagnostic testing for cystic fibrosis” by Handyside et al. One of a pair of papers that first used PCR to test embryos for genetic disease so that the disease free ones can be implanted.

So I looked at where preimplantation diagnostic testing was available in Lexington. There is an IVF clinic that does preimplantation testing–not at the UK hospital, they seem shy of anything slightly controversial. What was odd, is ZDL, a company that sells fertility products for people to use at home, or send in for analysis. Home sperm counts, insemination devices, and more. Quite surprising!

A was listening to Charles Darwin’s The Voyage of the Beagle, an incredible story of a round the world voyage of exploration that began in 1831 and continued for five years.

The end of the edition I was listening to had two appendices, the first being nautical stuff about the trip and the second being a reprint of the ship captain Robert Fitz Roy’s “Remarks with reference to the Deluge”. It’s basically a argument for literal biblical creationism by Fitz Roy, and it was pretty ironic to find it tacked on to the Voyage. I don’t know the which editions carried it, apparently the Voyage was published in several editions (wikipedia).

The strangest part of Fitz Roy’s argument was his attempt to describe a plausible way that the biblical Deluge, the business with the Ark and world covered by water, could have deposited the many layers of rock and sediment that compose the geological record and which he observed at locations around the globe. In his argument he mentions that the Library of Useful Knowledge, 1829 describes an experiment by Perkins that showed that at a depth of 3000 feet, water is compressed to 1/27 of its volume at the surface. Fitz Roy relates this to a sailor’s experience that in determining depth with a weight and a line, larger weights are required for deeper depths.

Fitz Roy then argues that sea water is very dense in the ocean depths, and that all sorts of objects immersed during the Deluge would sink to a depth where the water was dense enough to make then buoyant, different objects finding a different natural depth, and that these layers would be preserved as the Deluge ebbed.

This is quite remarkable. Water is known now to be nearly incompressible, around 2% at the bottom of the ocean, with the slight changes in ocean water density caused more by temperature and salinity than pressure. This means that Fitz Roy’s argument falls apart. More remarkable is that such a basic fact about water and the oceans was unknown in the 1830’s. Fitz Roy really believed there was a depth in the ocean where cannon balls float, accumulated from the world’s shipwrecks and banging around together in a layer where iron floats in the deep ocean.

With next generation sequencing technologies that have become available over the last two years, there is enough DNA sequencing machines that their combined capacity is, at a guesstimate, about 30,000 Gb per year.

How much DNA sequencing is that? Enough to sequence the genomes of 10,000 people per year. In other terms, enough capacity today to sequence every bacteria and virus species in a single year, or every the genome of every species on the planet in 300 years, bacterial genomes being small.

The sequencing technology is improving at a fast clip and I expect that in ten years or so it will be 1000X better (faster and higher capacity sequencing machines). So in ten years, it will be possible to sequence every species on the planet in a single year.

Scanners can be used for macro photography, at least the ones with a great depth of focus. The scanners with CCD sensors tend to have a good depth of focus while the thin scanners with a CIS sensor can only focus on objects a mm or two from the glass.

I collected a list of scanners recommended for macro photography.

from here:
EPSON Perfection 3170
from here.
Microtek ScanMaker X6 EPP
Artec AM12S, AM 2400-U Pro
Epson Expression 836XL
3D Pro Scanner
Memtek Memorex SCF 9360P 3D
from here:
Epson Perfection 1240U Scanner
Epson Perfection 1200U

Checking specs, these also look well suited:
Epson Perfection 2480
Epson Perfection V300
Epson Perfection 2450
Epson Perfection 3490

Two recent posts on the theoildrum.com look at the question of wind power. The first post looks at the how practical it is to fit wind power into the US power grid. The main problem with wind power are that sometimes the wind doesn’t blow, so the power grid has to have excess generating capacity to meet the demand. Critics say that because of the ‘no blow’ times, wind power can’t replace base capacity, so even if lots of windmills get installed, the US still needs all the coal, nuclear, and gas power plants.

The theoildrum.com article considers the problem and concludes if wind farms are spread out and high capacity transmission lines are built to pool the power wind farms should be able to provide base power at about a quarter of the total installed windmill capacity. The power grid should be able to accommodate somewhere between 25% to 50% of US power coming from windmills.

‘Smart grid’ capability, having devices like A/C that temporarily shut off when demand is too high is also an option. Wind power is also nicely complementary to hydroelectric generation, as a dam stores power and the turbines can be spun up and down quickly as average wind strength varies.

The other theoildrum.com article looks at the cost of wind power and at whether anything limits the prospect of building lots of windmills today. There appear to be no resource that constrains windmill production. Today windmills are cheaper than anything but coal in the US, and modest carbon taxes would make wind power the cheapest power source:

It is time to build windmills, and lots of them!

A child inherits half each parents genes, and for any individual gene a sibling can either inherit the same or the other allele. Averaged out over the genome, siblings have the same allele for about half the genes.

But genes don’t assort independently of each other, genes are found on chromosomes and due to this inheritance of gene variants is lumpy, a parent has a pair of each chromosome and siblings either get the same chromosome or each gets one of the parent’s pair. This happens randomly for each of the twenty-three pairs of human chromosomes. If humans had only two chromosomes then one forth of the time two siblings would inherit the same pair (out of the four) and be genetically identical. Because humans have twenty-three chromosomes this is vanishingly unlikely. But while on average two siblings have the same gene variant half the time, there is actually a distribution centered at 50%, and siblings can inherit more or less than 50% of same alleles. So how likely are siblings to be 30% or 40% related?

Here’s what the distribution looks like. There are a few more wrinkles to consider. First off, the chromosomes recombine before they assort, so each chromosome a child inherits is a combination of the parents chromosomes. It turns out that in humans, recombination happens more than once per chromosome, about thirty-three times (The Human Genome Project
By Necia Grant Cooper, p31). I include this in my model. I don’t consider the chance that a parent will have two copies of the same allele for a gene, or how recombination is more likely at particular places along a chromosome, or that genes are clumped together in certain chromosomal regions. Here’s the distribution I find:

So siblings don't always share half their gene variants, but rather have a modest chance or being a bit more or a bit less related.