Correlation or causation or neither?

Andras Pellionisz is an interesting fellow. He appears to have been a productive researcher, who then saw the chance to make some real cash and apparently started up a couple of companies.

Of late, however, he seems to be on a megalomaniacal tirade of sorts, setting himself up as the greatest thing in science since Crick (whose 'dogma' he (Pellionisz) has done away with), with his "establishment" of his "Principle of the Recursive Genome" (I won't even mention his malicious, self-aggrandizing historical revisionism on junkDNA...).

In short, Pellionisz claims that what we used to refer to as 'genes' interact recursively with what we used to call 'junk DNA' to produce structures that can be described via fractal mathematics.

He 'predicted' in 2006 that because of this 'principle', the Purkinje cells (P-cells) in cerebellums of fish (or other organisms with brains) of 'more recently' developed status, having larger genomes, will have greater branchings in their dendritic networks:

One of us (AJP) has previously proposed that fractal processes associated
with DNA are in a causal relation to the fractal properties of organelles such
as P-cells (FractoGene, 2002, patent pending). [let us ignore for now the odd
depiction of a neuron as an 'organelle'] This fractal postulate
predicts that the dendritic arborization of P-cells will be less complex in
lower order vertebrates.

Well, OK. A prediction that Fugu will have less arborized dendrites than mouse and human because ... well, let's see...

The prediction can be tested by systematic comparative neuroanatomy of the
P-cell in species for which genome sequences permit inter-species comparison.
The Fugu rubripes (Fugu), Danio rerio (Danio) and other species are lower order
vertebrates for which genome sequences are available and tests could be
conducted. Consistent with the fractal prediction, P-cell dendritic arbor is
primitive in Fugu, being much less complex than in Mus musculus and in Homo
sapiens. Genomic analysis readily identified PEP19/Pcp4, Calbindin-D28k, and
GAD67 genes in Fugu and in Danio that are closely associated with P-cells in
Canis familiaris, Rattus norvegicus, Mus musculus and Homo sapiens. Gene L7/Pcp2
exhibits strongest association with P-cells in higher vertebrates. L7/Pcp2 shows
strong protein residue homology with genes greater than 600 residues and
including 2-3 GoLoco domains, designated as having G protein signaling modulator
function (AGS3-like proteins). Fugu has a short gene with a single GoLoco
domain, but it has greatest homology with the AGS3-like proteins. No similar
short gene is present in Danio or in Xenopus. Classical L7/Pcp2 is only detected
in higher vertebrates, suggesting that it may be a marker of more recent
evolutionary development of cerebellar P-cells. We expect that a new generation
of data mining tools will be required to support recursive fractal geometrical,
combinatorial, and neural network models of the genomic basis of morphogenesis.

OK, great. Looks like he's onto something. Right?
Now, however, one finds a hyperbole riddled webpage announcing:

UPDATE (14 September, 2007) Question-marks raised by the "Fugu prediction
paper" [the paper mentioned above] have met tentative support. The
correlation - contrary to blogs whose ideology would prefer otherwise -
is further supported, there is no contradiction, no gaps, whatsoever.
Peer-reviewed presentation of evidence from independent sources will follow
whenever they are ready.

Please note the inconsistency between the first and second sentence - in the first, he states that 'question marks' in his Fugu paper have found "tentative support", but the second indicates that there is near certainty.

Anyway, the commentary there is hard to follow, but there are some figures apparently from the paper claimed to have supported his Fugu 'questionmarks' and their legends:

[original legends] Fig. 4. Sketch of the emerging field of comparison of
the complexity of the dendritic trees of P-cells, their genomic analysis,
calling for biophysical synthesis. Insert B shows the P-cell in the Fugu
rubripes (B is courtesy of Székely33), in which the genome size is 0.37
Gigabases. C will show the P-cell in Danio rerio (as it becomes available,
according to studies at an early stage to exhibit an interim complexity) in
which the genome size is 1.56 Gigabases. D shows the dendritic arbor of the
P-cell in the mouse (genome size is 2.6 Gigabases). Insert D is
fluorescent-stained photo, courtesy of Prof. Helen Blau40. E shows a
computer-reconstruction of the P-cell in the guinea pig21. The genome size in
the guinea pig is not known to date, but its sequencing was slated (at Broad
Institute and MIT) among other species. Insert H shows the P-cell of the
human27. The genome size in the human is 3.1 Gigabases.

[updated legends - 14 September, 2007] Fig. 4. Sketch of the emerging
field of comparison of the complexity of the dendritic trees of P-cells, their
genomic analysis, calling for biophysical synthesis. Insert B shows the P-cell
in the Fugu rubripes (B is courtesy of Székely33), in which the genome size is
0.37 Gigabases. Question-marks in C will be replaced by finalized results of
already known preliminary studies exhibiting an interim complexity, in which the
genome size is 1.56 Gigabases. D shows the dendritic arbor of the P-cell in the
mouse (genome size is 2.6 Gigabases). Insert D is fluorescent-stained photo,
courtesy of Prof. Helen Blau40. E shows a computer-reconstruction of the P-cell
in the guinea pig21. The genome size in the guinea pig is not known to date, but
its sequencing was slated (at Broad Institute and MIT) among other species, and
now preliminary sequencing results project the the genome size right in the
predicted range***. Insert H shows the P-cell of the human27. The genome size in
the human is 3.1 Gigabases.

Yes, that is pretty much exactly how it appears on his website (one of apparently a dozen or so that he maintains).
So, they show the P-cell in a Fugu with like 3 dendrites, and 'more recently developed' P-cells with greater arborization (hundreds of branches).


So this is where I get lost - the recursive genome claims appear to indicate a 're-visiting' to the genome areas associated with a particular structure or process by proteins/RNAs with the end result being a fractalicious outcome.

His 'support' appears to be a correlation between genome size and arborization of P-cells in the cerebellum.

While this may well hold true for P-cell arborization in a general sense, what we do NOT see is any indication of 'recursivity' at all - what the 'support' paper shows is that the 'more recently developed' organisms have more/larger genes associated with their P-cells:

L7/Pcp2 showsstrong protein residue homology with genes greater than 600
residues andincluding 2-3 GoLoco domains, designated as having G protein
signaling modulatorfunction (AGS3-like proteins). Fugu has a short gene
with a single GoLocodomain, but it has greatest homology with the
AGS3-like proteins. No similar short gene is present in Danio or in
Xenopus. Classical L7/Pcp2 is only detectedin higher vertebrates,
suggesting that it may be a marker of more recentevolutionary development of cerebellar P-cells.

And that was from HIS OWN paper! Implicit in his junk DNA claims is the notion that the recursivity is a function OF the junk DNA (and so it is not junk). To prop this notion up, he has made a great deal of noise about the fact that one can find stop codons in junk DNA. It has been pointed out to him that you can find ALL of the codons for ALL amino acids (including MET - start codons) in 'junk DNA' , too. But he insists that stop codons in junk DNA are special and supportive of his claims. Somehow - he won't really say how, just that they are.

So is the larger genome - more 'junk DNA' anyway - of a 'more recently' evolved oganism the cause of the greater arborization in their P-cells and thus Pellionisz's recursive genome principle is real and true? Or is greater arborization unrelated to genome size directly and the degree of arborization has a different cause - maybe arborization is influenced by inductive processes* that have little if anything to do with recursivity?


Like anyone envisioning him or herself as a paradigm-busting visionary - who also has a product to sell - Pellionisz seems to make reckless extrapolations and to ignore or downplay potentially disconfirming evidence (none of the 93 citations in his 'Principle' paper, for example, seem to even mention indiction).


Now, about that junk DNA revisionism of his.....


UPDATE:
While perusing the web for more interesting info on Pellionisz's claims, I came across his foray onto the Panda's Thumb last year. As I mentioned, Pellionisz cites as "evidence" for his claims the observation that Fugu has a smaller genome compared to 'higher' vertebrates and also has less arborized P-cells. Andrea Bottaro asks him to look at the lungfish, whose genome is some 10 times the size of our own, to see if they have 10X the arborization that human P-cells have.
Pellionisz's response was to engage in some ego-boosting and ranting but not once did he even mention testing his claims as Bottaro suggests. In fact, when it came up again later, Pellionisz only mentioned the guinea pig genome, which he declared to fulfill one of his 'predictions'. Apparently, he is afraid that the lungfish will negate the predictive power of his 'theory' (which he claims is true). Later, when pressed again, he punted, claiming that he believes that their larger genomes probaly have to do with metamorphosis and have nothing to do with the cerebellum... Cute, but it would clearly violate his claims re: the P-cells in Fugu, guinea pig and human. Not to mention the fact that lungfish do not undergo metamorphosis.




Later, it is pointed out that lots of people were research in junk DNA before he and his hero M. Simons came along, and that their claims of martyrdom were just to generate sympathy for their cause. Pellionisz responds to that by ignoring the fact that junk DNA research had been going on all along and instead merely trying to gain more sympathy by mentioning that Simons has a 'junk DNA-related' disease. Pellionisz also refers to promoter regions as "junkDNA", implying that research on such areas had 'also' been neglected. Such regions have been not only hypothesized but in fact known about for decades.

Also, when he was reminded that things like promoters had been discovered prior to Simons' work and that research into junk DNA had been going on all along, he disissed it by quoting a section from Brenner's Nobel speech in which he characterized junk DNA as useless. The speech was given in 1985.
Later in the comments, we see this:
" Perhaps he feels I have no right to comment, since my Ph.D. is in sociology, not science proper. But I must say, if my work in the sociology of science had played as fast and loose with historical accuracy as Dr. Pellionisz’s does, I would have been booted out of the program before I got my MA."

Indeed.


UPDATE UPDATE:

Pellionisz later (in the PT thread linked above) states:

"Since Malcolm (Simones, Pellionisz's 'partner') is widely known to be affected by a “Junk DNA disease” himself, after having pinned down that “junk was anything but” he is motivated having already spent two decades to finding out the best approach to what might exactly be the function that Darwinian theory actually required. "

More historical revisionism from Pellionisz. Simons published ZERO papers having anything to do with junk DNA or junk DNA diseases or anything until about 10 years after Zuckerkandl has identified promoters and hypoethesized about functionin junk DNA. Amazing.




*I am aware that only a few of the returns on those searches deal directly with the topic of arborization of P-cells, my point is there ARE other potential explanations which one can find by doing a little searching.