Every now and again I read something which fills me with amazement about the stubborn, relentless persistence of the human mind.
Researching the structure of dolomite, a variation of calcium carbonate which incorporates magnesium and which, because of its great strength I want to use more of in my next fresco plaster, I was intrigued to learn that scientists had recently set about looking for signs of life in Triassic dolomite amber. Amber is a beautiful material surviving the ravages of time, and in which insects, pollen and various other artefacts of micro-eco systems are found trapped and perfectly preserved, virtually for all time. But it was the way they did it which astounded me – they searched seventy thousand droplets individually, to find the ones they wanted – numbering exactly three:
Undaunted by knowing they were looking for a needle in several acres of haystacks, they gathered “thousands of droplets of amber no more than 6mm long from outcrops on the Dolomites, then spent two years screening each tiny piece for any plant or creature they might contain. If you think this must have been a painfully tedious experience, you think right.”
Here’s the full story, from the History Blog:
Scientists have found two gall mites and a fly preserved in 230-million-year-old Late Triassic amber from the Dolomite Alps in northeastern Italy. These are the first arthropods found in amber from the Triassic. Although arthropods appear in the fossil record starting in the Early Cambrian period around 540 million years ago, these are the oldest arthropods ever discovered trapped in amber, 100 million years older than prior examples.
“Before preparation, one of the tiny flecks of amber, about 1 millimeter in diameter, wafted onto the floor of my lab,” [David Grimaldi, curator of invertebrate zoology at the American Museum of Natural History in New York,] recalled. “Alex Schmidt and my assistant who did the prep, Paul Nascimbene, spent about three hours on their hands and knees with flashlights. I don’t know how, but they found the speck on the floor hidden in the corner between two lab benches. It was a nerve-wracking time.”
It’s a good thing they were that dedicated to the task, because out of the 70,000 droplets of amber they analyzed for inclusions, three of them contained the groundbreaking arthropods: one partial midge fly (Diptera) about the size of a head of a pin, and two new species of gall mites. The amber had preserved the head, antenna, four legs and some fragments of the body of the midge fly which taken together were 1.5-2 millimeters in size.
The mites were complete and far tinier. One of them, Triasacarus fedelei, is shaped like a worm and is 210 microns long. Due to its body shape and differences in the structure of the mouth, researchers believe it may be an ancestor of modern gall mites. The second mite, Ampezzoa triassica, is even smaller at 124 microns in length. Its shape is more typical of modern gall mites.
The amber preserved these creatures so well that scientists could examine them in microscopic detail, allowing them to identify minute characteristics like tiny waxy filaments on the surface of Triasacarus fedelei’s body that are thought to have helped protect them from predators, parasites and the elements. Both mites have two pairs of legs like present-day mites.
One way in which they’re different from their modern relatives is that the Triassic mites must have fed on conifers, since it’s conifer resin that hardens into amber. Today 97% of Eriophyoidea, the group gall mites belong to, feed on flowering plants, which means the mites changed their eating and living habits entirely when the new plant species came on the scene about 140 million years ago.
Some kinds of perseverance deserve special recognition and this qualifies for sure. Life seems to arrive with persistence. We see grass persist long enough to break concrete to pieces, and all of us persisted in learning to walk despite bruising our knees daily. The planet keeps trying to clean itself, and we keep trying to mess it up again. Life doesn’t give up, and it hardly matters whether that’s a quality of physics or of consciousness since either way, it is part of whatever makes us alive.
In other news, it was announced this week that the number of planets in the Milky Way has been grossly under estimated. Years ago in high school we were told planets were bound to be unusual occurrences and as far as anyone knew, we might well be on the only inhabitable one in the universe. This was pretty dispiriting stuff – but as the decades rolled on, scientists’ eyes became bigger through telescopes and star-wobble detectors until they could see that giant planets, orbiting very quickly, weren’t all that uncommon after all.
Even so, it was common to hear that life was bound to be a rarity – considering the large number of factors that had to be just-so, and considering that, for all we know, rocky inhabitable planets were sure to be few and far between. Man was an exception and probably the smartest thing in the universe. And remember, TV remotes were still attached with long cables, and cigarettes were cool as hell.
Impartial time rolled on, oblivious to the welter of opinions for and against life elsewhere. Now things have changed. Of course they haven’t at all – but our opinions sure have:
..A fresh analysis of data from NASA’s Kepler mission, which launched in 2009, according to new research presented at the annual meeting of the American Astronomical Society in Long Beach, California, indicates:
“..we found the occurrence of small planets around large stars was underestimated,” said astronomer Francois Fressin, of the Harvard-Smithsonian Center for Astrophysics in Cambridge, Massachusetts.
To find planets, Kepler stares at a patch of sky in the constellation Cygnus, made up of about 150,000 stars. The space telescope detects potential alien worlds by watching for telltale dips in starlight created when planets pass in front of, or “transit,” their parent stars.
Using their own independent software for analyzing Kepler’s potential planet detections, Fressin and his colleagues estimate that about 17 percent, or one in six, of all the sunlike stars in the Milky Way host a rocky planet that orbits closer than the distance at which Mercury orbits our own sun.
Since the Milky Way is home to about a hundred billion stars, that means there are at least 17 billion rocky worlds out there. (See Milky Way pictures.)
When the team expanded their search to Earth-size orbits or larger, they found that half of all sunlike stars may host rocky planets.
“Every time you look up on a starry night, [nearly] each star you’re looking at has a planetary system,” Fressin said.
Impressive! But as the article goes on to say, Rocky planets are just a fraction of the total number:
A study of the number of potential worlds orbiting M-dwarfs—faint stars smaller than our sun that make up the vast majority of the stellar population—suggests our galaxy may be home to at least a hundred billion planets overall.
“Based on our calculations we are showing that there is about one planet per star, and that gives us a total of about a hundred billion planets throughout our galaxy,” said Caltech planetary astronomer John Johnson.
So now the debates turn to whether those planets can support life. But here’s the thing: what’s so unusual about our star? And why should our planet be exceptional? I’ve read very convincing testimony regarding UFOs from the earliest pilots to astronauts on moon missions, hard-headed realists all. But believe them or not, I think most people can’t avoid the steadily dawning idea that as our eyes open, this 14 billion year old Universe must be absolutely teeming with life.
The other gnawing idea is that as late arrivals, we with our cherished worldviews rapidly unravelling in all directions, our insecure need to be smarter than the next guy, our wars and killings and stockpiles of Hellish weapons, our willingness to threaten this terrifying prospect against our own species as a daily norm, while claiming to fight terror, and our dogmatic fundamentalisms, are most likely still on the very lowest rung of the intelligent life ladder: a ladder which could stretch a very, very long way indeed.
This is an article I wrote a year ago which ranks near the top in hits, and is still one of my favourites: if anything, it shows the marvellous long-term precision involved in molecular engineering, a challenge which still defeats our keenest intellects. Here again then, are:
Links to the following five research articles are given at the end, for those who wish to delve more deeply, but the gist of each remarkable point is as follows:
(1) The codon bases have a non-random correlation with the kind of amino acids which they code for. In other words, the first out of the three letters of a codon relates to the kind of amino acid the codon stands for, giving the language a deeper meaning.
This undoubtedly helps the error checking machinery just as the quickest kind of computer program to debug is one in which variable names include a consistent reference at the start classifying it as holding a date, integer, text, array, database value, or whatever. Without this, you can’t debug it at a rapid pace because every variable name needs to be consciously checked.
So it seems the codon table was assembled using the rules of good programming:
“Now class, what does this circular chart represent..? Yes – Mr Dawkins? Tsk tsk! It is most certainly not a dartboard. Ah yes.. you there at the back – Dennett, is it..? Good God, no, it is not a roulette wheel either! Has even a single one of you done your homework?!”
(2) The greater or lesser effect of errors (mistranslations) when the DNA is translated is called the “load on the code” and is minimised by its current arrangement to such an extent that only 3 in 100,000 other possible mappings would have a safer error rate, but might also have some deleterious effect on the overall DNA function, as a single change in the codon mapping would cause huge atomic changes throughout the length of the three billion base pair system, as well as requiring modifications to all of its interpretation and duplication machinery, which seems to be geared up for this specific arrangement.
It is worth noting that this statistic already assumes that all 100,000 alternatives already have the advantage of the type-significant first letter of the codon (detailed in (1) above) which is already known to have an error-reduction benefit. Therefore if one were to include all the truly random arrangements – where the first letter was not weighted towards codon relevance – the disproportion would be vastly greater than 3:100,000.
To give you an idea of the scale of the overall odds when starting from scratch, instead of considering only those mappings with a known advantage, the number of completely random alternatives less effective at error-reduction (by virtue of a random first letter instead of a significant one) would be 64*63*62*61*60..*45 which if we forget about start and stop codons I work out as 47,732,870,256, 486,900,000,000, 000,000,000,000. Subtract 36,267,774,588,438,900,000 representing the 3 results per 100,000 estimated as being more fault-tolerant than the current design and you get a worse-performing set of 47 * 10 to the power 33. This means if you had a trillion planets around every star in the Universe, you could try a different arrangement on every planet, and get a highly fault tolerant system only once.
To put it another way, you could try out a different mapping system on a different planet circling every known star in the Universe, each and every day for 3 billion of our years and stand a chance of getting a better mapping precisely once. And in that day for each test version you’d need to create a complex life form from scratch, and subject it to every imaginable adverse circumstance – seasonal, predatory, infectious, organ failure, sensory development etc., and gauge its reproductive success in 24 hours before throwing it out and organising a new one.
Even if a better mapping was arrived at, it would still need to be evaluated as to its effect onspeed of protein assembly or the combined molecular effect of the billions of changes throughout the length of the entire chromosomes. All things considered I’m kind of backing the system we have now: it must be hugely error-tolerant to allow life forms to remain unchanged for 200m years (sharks).. 160m years (Isopods).. or certain insects presumed to have formed during the Silurian period 400m years ago, during all of which time the codon mapping had to have remained constant.
Anticodon-amino acid enrichment observed in ribosomal structures is correlated with the canonical genetic code. The distributions of the average enrichment value of codon (Left) or anticodon (Right) for 1,000,000 random codes are shown with the arrow pointing to where the canonical code stands. The correlation analyses were performed using (A) a specific subset of amino acids that is optimal for the canonical code; (B) all amino acids except cysteine; and (C) a subset of amino acids optimal for each code. A significant proportion of randomized codes have a lower average enrichment than the canonical genetic code in ribosomal anticodon-amino acid enrichment. (www.pnas.org)
As Richard Dawkins says in his analysis of the laryngeal nerve (specifically of the giraffe but in general, also in the human design) and its evolution from the fish model, the DNA can’t go back and start again. Ever since the embarrassing “backwards retina is bad design” argument was discredited, and the tonsils turned out to be not absurd inconveniences but immune system sentinels guarding the main entrance, and even the long-ridiculed appendix has been shown (SciAm, March 2012) to be a little farm breeding protective bacteria, so that those who have had it removed tend to get nasty complications much more frequently than those who still have it, and the much-derided “junk DNA” turns out to be full of structural code and even jumping genes (retrotransposons) containing changes to the brain’s schematics), critics are hard put to find any flaw in biology.
That doesn’t mean they stop looking – like a financially overstretched holidaymaker desperately looking for any reason to avoid paying the bill! The odds against life may indeed be staggering, but the mechanisms themselves use every bit of cunning engineering imaginable to overcome them. The cheery catchphrase “God just represents the shrinking area we don’t understand” seems to be turning into “OMG, the area we thought we understood seems to be getting smaller.”
160 million year old genetics — yes, still working fine!
(3) The coding system is given further weight by the discovery that within the ribosome, anticodons are enriched near the areas relative to their function, to a level such that the probability of this being a random setup is minuscule. Not minuscule the way a likelihood of 6.9 is a very small step away from an impossibility level of 7, but less than .0000000000000000001 %.
In other words, the ribosome behaves as if it’s already geared up and ready to work with the existing code – and yet is assumed to be one of the most ancient parts of the whole DNA engine.
..beautiful graphic from http://ionsource.com/virtit/VirtualIT/aainfo.htm
(4) Speaking of ribosomes, they are so well structured that even when the component parts are broken down by chemical catalyst into long molecular fragments and more than fifty different proteins – they reform into a functioning ribosome as soon as the divisive chemical forces have been removed, independent of any enzymes or assembly machinery – and carry on working. Design some machinery which behaves like this and I personally will build a temple to your name!
Maybe he simply threw the dice 10,000,000,000,000,000,000,000,000,000,000,000,000,000,000,000,000,000,000,000,000,000,000,000,000,000,000 times. But who swept up all the rubbish? Image: http://faithforthinkers.blogspot.com/
(5) Last but not least, some new evidence that the DNA is a toolkit with non-random (purposeful) correlations between similar structures. This seems intuitively necessary but it occurs in completely unrelated proteins. This is evidence that the same kind of engineering underlies separate biological components when similar functionality is needed. With a smallish toolkit of 20 amino acids, you would expect random constructs to be full of illogical or redundant features that other components somehow had to compensate for.
What you actually find is that protein structures are so perfectly aligned they immediately spring from their two dimensional string into a three dimensional shape (complete with cross-member welds, sprung sections, and fully integrated behaviour such as expansion and contraction which all rely on a perfectly balanced three dimensional design) via the laws of physics, chemistry and electrical engineering, and finally (by their actions) fulfilling laws of biology.
They are so hard to understand because understanding first requires mastery of all these schools of thought as well as the ability to think in terms of three dimensional electrical forces. The speed at which they snap into shape and the rate at which they carry out their function is frightening (in the former case, reducing a possible random trial of arrangements measured in multiples the age of the Universe to a few milliseconds, and in the latter case, performing as many as ten million actions per second in the case of some enzymes).
This ingenuity and success does not randomly result from the three dimensional state – at this stage it is already too late to modify the strip of amino acids to correct for errors – but must be planned in advance within the original two dimensional strip of DNA. This is a staggering achievement by any standards.
1. Non Random Codon – Amino Acid Mapping
For the first time it is shown that each of the three codon bases has a general correlation with a different, predictable amino acid property, depending on position within the codon. In addition to the previously recognized link between the mid-base and the hydrophobic-hydrophilic spectrum, we show that, with the exception of G, the first base is generally invariant within a synthetic pathway. G– coded amino acids show a different order, being found only at the head of the synthetic pathways.
The redundancy of the nature of the third base has a previously unrecognised relationship with molecular weight. The bases U and A (transversions) are associated with the most sharply defined or opposite states in both the first and second position, C somewhat less so or intermediate, and G neutral. The apparently systematic nature of these relationships has profound implications for the origin of the genetic code. It appears to be the remains of the first language of the cell, predating the tRNA/ribosome system, persisting with remarkably little change at a deeper level of organisation than the codon language
Molecular representation of a ribosome: 250,000 atoms. The idea that a superhuman intelligence can understand it actually gives a target for human intelligence to aim for, since if it was a completely random design, there would be no point in studying its particular arrangement with the hopes of understanding its logical function, since any other random arrangement might be equally effective
2. Non Random Fault Tolerance in Codon Mapping
The average effect of errors acting on a genetic code (the change in amino-acid meaning resulting from point mutation and mistranslation) may be quantified as its ‘load’. The natural genetic code shows a clear property of minimizing this load when compared against randomly generated variant codes.
Here then, we ask whether this ‘historical’ force alone can explain the efficiency of the natural code in minimizing the effects of error. We therefore compare the error-minimizing ability of the natural code with that of alternative codes which, rather than being a random selection, are restricted such that amino acids from the same biochemical pathway all share the same first base.
We find that although on average the restricted set of codes show a slightly higher efficiency than random ones, the real code remains extremely efficient relative to this subset P = 0.0003. This indicates that for the most part historical features do not explain the load- minimization property of the natural code. Once mistranslational biases have been considered, fewer than four per 100,000 alternative codes are better than the natural code.
3. Non Random Anticodon Toolkit in Ribosome
The establishment of the genetic code remains elusive nearly five decades after the code was elucidated. The stereochemical hypothesis postulates that the code developed from interactions between nucleotides and amino acids, yet supporting evidence in a biological context is lacking. We show here that anticodons are selectively enriched near their respective amino acids in the ribosome, and that such enrichment is significantly correlated with the canonical code over random codes. Ribosomal anticodon-amino acid enrichment further reveals that specific codons were reassigned during code evolution, and that the code evolved through a two-stage transition from ancient amino acids without anticodon interaction to newer additions with anticodon interaction. The ribosome thus serves as a molecular fossil, preserving biological evidence that anticodon-amino acid interactions shaped the evolution of the genetic code.
Analysis of ribosome structures, shown on the left, from four different species revealed a non-random affinity between anticodon-containing RNA triplets and their respective amino acids, shown on the right). Credit: David Johnson, Salk Institute for Biological Studies
4. Self-Assembling Ribosome
Ribosomes are complex cell organelles consisting of ribosomal RNA (rRNA) and a number of different protein molecules. One of the major questions one can ask about the ribosome is how these component molecules are assembled into the final organized structure. As described in the preceding article, we have recently succeeded in developing a reconstitution system that, under well-defined conditions, produces physically and functionally intact 30 S ribosomes from free 16 S RNA and free 30 S ribosomal proteins from E. coli (Traub and Nomura, 1968). Reconstituted 30 S particles synthesized under optimal conditions were found to be indistinguishable from the original 30 S ribosomes with respect to their sedimentation coefficient, their protein composition, and their capacity to bind aminoacyl-tRNA and to synthesize protein in response to messenger RNA (mRNA). Furthermore, this total reconstitution system proved to be independent of the presence of additional macromolecular structures…
5. Toolkit Structures in Protein Amino Acid Sequences
Unrelated proteins with high percentage identity. Hemoglobin β-chain (pdb code 1hds chain b, ref. 38, Left) and cellulase E2 (pdb code 1tml, ref. 39, Right) have 39% identity over 64 residues, a level which is often believed to be indicative of homology. Despite this high degree of identity, their structures strongly suggest that these proteins are not related. Appropriately, neither the raw alignment score of 85 nor the E-value of 1.3 is significant. Proteins rendered by rasmol (40).
Why reinvent the wheel? ..Exactly