Jeff Tomkins has published another paper in Answers Research Journal on the DNA similarity between humans and chimpanzees. Now there is quite the story behind this paper, and I’m sure I’ll address that in another post. For now, I just want to concentrate on the very last paragraph:
“And second, the majority of flow cytometry studies of chimpanzee nuclei along with the cytogenetic analysis of chromosomes indicate a genome size difference of about 8%, with the chimpanzee genome having a significantly larger amount of heterochromatic DNA compared to human (Formenti et al. 1983; Pellicciari et al. 1982, 1988, 1990a, 1990b; Seuanez et al. 1977).”
I’ve always been a bit skeptical of claims that the chimpanzee genome is bigger than the human genome, simply because I have the latest versions of the genomes on my hard drive. A chromosome by chromosome breakdown of each genome shows that it is actually the chimpanzee genome is slightly smaller than the human genome. There is also a handy website – http://www.genomesize.com – that collects references for the sizes of hundreds of genomes.
It’s important to point out that the DNA content of organisms – although measured in picograms – is very rarely “weighed” in the sense of putting it on a set of scales. The method used in these studies involves staining the DNA with a special compound that glows when excited by ultra-violet light. These studies are really only reporting how much a certain sample glows more than or less than a given standard. Older studies tended to use a value for the human genome of 7.30pg to calibrate their results, while newer studies tend to use a value of 7.00pg.
The citation given on this website for the standard human genome is from a paper in 2005 written by Awtar Krishan et al. This study measured the genome size of a whole bunch of animals from the Miami Metro Zoo and used two samples from a human male to calibrate the results. There were three chimpanzee samples measured: 7.22pg (for the female), and 6.77pg (for two males).
So, straight away we can say that at least in this study, the chimpanzee male genome is about 3.4% smaller than the human male genome. And since we know roughly how big the X and Y chromosomes are in humans (156 million base pairs and 57 million base pairs respectively), and how much they would weigh in picograms, we can calculate the weight of a human female diploid cell: 7.10pg. So it appears that the female chimpanzee genome is about 1.7% larger than the human female genome.
So let’s look at the papers that Tomkins cites to support that final paragraph:
Formenti et al. 1983
The actual title of this paper is “Variazioni del Contenuto Nucleare in DNA Negli Hominoidea” and was published in the Italian-language journal “Antropologia Contemporanea“. A search online for the abstract for this paper yielded precisely zero results. The paper is cited by genomesize.com as giving a result of 3.63pg, but we do not know if this was for a male or a female chimpanzee. If it was for a male, then the chimpanzee is 3.7% larger. If it is for a female, then the chimpanzee is only 2.2% larger. Given our result above, it seems a little more likely than not that this was from a female chimpanzee.
Pellicciari et al. 1982
This paper can be found here, although unfortunately for most lay-people, it is behind a paywall.
The chimpanzee figure given in this paper is 8.03pg against a standard human genome size of 7.30pg, which is a full 10% larger. But things get a little interesting when you consider:
“Table 2 reports, for each species examined, the Feulgen-DNA contents (in pg and in percentage as compared to man) and the morphological data of the karyotype (chromosome number, 2n and fundamental number, FN) drawn from the literature (De Boer, 1974; Chiarelli et al., 1979). Data so far unpublished are marked with an asterisk. Previously published data have been recalculated on the basis of the Feulgen-DNA content of the control species included in the corresponding lots of preparations.”
And since chimpanzee is one of the entries not marked with an asterisk, the authors appear to be citing a previously published figure, presumably from Chiarelli et al., 1979 (Comparative Karyology of Primates). Delving into this source, we see that it is actually a symposium, featuring previously published papers:
“A general account of the literature available on primate chromosomes up to 1972 has been collected in the last part of this book.” (p27).
Taken in combination with the following quote from the original paper (Pellicciari et al., 1982):
“Finally, in Hominoidea (Fig. 1 e), Pongidae have a variable Feulgen-DNA content, the value being higher than in man. This last finding, previously published by us (Manfredi Romanini, 1972) has been recently confirmed by Seuanez et al. (1977) by microinterferometric studies on spermatozoa, even though the content sequence observed by this author is different from ours (Homo < Gorilla < Pan < Pongo in Manfredi Romanini, 1972; Homo < Pan < Pongo < Gorilla, in Seuanez et al., 1977).”
… and it appears that the original source for this figure actually dates back to 1972. Chasing that source down, we see the following results (in arbitrary units, which are then scaled to picograms using the standard human genome size at the time):
Pan troglodytes: 13.60 ± 4.60; Homo sapiens: 12.36 ± 0.11
These standard deviations are enormous! On these numbers, it’s possible that the chimpanzee genome could be 27% smaller than the human genome.
Pellicciari et al. 1988
This abstract can be found on PubMed, but I cannot find the full text online anywhere. In regards to Tomkins’ citation of it, the title is already a huge cause for concern: “Genome size and constitutive heterochromatin in Hylobates muelleri and Symphalangus syndactylus and in their viable hybrid“. The title makes no mention of measuring chimpanzee DNA, and nor does the abstract: “Genome size was measured […] in six species of the family Hylobatidae and in a hybrid of the gibbon (Hylobates muelleri) and siamang (Symphalangus syndactylus)“. However, genomesize.com comes to the rescue, and according to its citation of it, the figure used in this paper was 3.63pg (against a human standard of 3.50pg). Given that it is near certain that no new measurement of the chimpanzee genome took place in this study, it seems more likely that this paper cites a previously published figure – and based on a comparison of the authors in both papers, it’s likely they are citing Formenti et al., 1983.
Pellicciari et al. 1990a
This paper is also behind a paywall, but the abstract can be found here. This paper uses flow cytometry to measure the genome size and the following results were obtained:
Pan troglodytes: 7.85pg ± 0.40pg; Homo sapiens: 7.30pg ± 0.35pg
So again we have reasonably large standard deviations in the samples, but taken at face value, the chimpanzee genome is 7.5% larger than the human genome.
Pellicciari et al. 1990b
This paper is also on PubMed here, and again from the abstract, it is quite clear that this paper does not perform a new measurement of the chimpanzee genome. “Measurements were performed by microfluorometry on […] man, gorilla and mouse“. Heterochromatic DNA was also measured in “man, gorilla and mouse“. Karyotypes were stained in “man, gorilla and mouse“.
Seuanez et al. 1977
Again, this paper is behind a paywall. This paper does not give any quantitative measurements, however, here I have reproduced an original graph from this paper. The method of measuring genome size here involves weighing the spermatozoa – “Total Dry Mass” (“TDM”), extracting the DNA, and then weighing the remainder – “Dry Mass After Extraction” (“DMAE”). The difference of course being the “Dry Mass of Extracted DNA” (“DNA-DM”).
In this graph, the small circles represent human DNA, while the small squares represent chimpanzee DNA. As you can see, one human data point is clearly higher than both chimpanzee data points and the other human data point is clearly lower than both chimpanzee data points. Given such high variance in the results and the fact that the authors did not publish the actual figures behind he graph, it is difficult to draw anything useful from this paper (other than the fact it conflicts significantly with Pellicciari et al., 1982‘s ordering of primate genome sizes, and could then count as evidence that the true measurement taken in that paper would be at the lower end of the range).
So where does this leave us?
Tomkins makes an impressive list of six papers in support of his position. Two of these can be discarded instantly – Pellicciari et al., 1988 and Pellicciari et al., 1990b – simply because they are obviously referring to previously published results rather than taking new measurements.
We have Pellicciari et al., 1982, for which the primary source is actually a paper from 1972, which claims a 10% size difference, but with an enormous standard deviation – so high that the human genome could quite easily be larger than the chimpanzee genome. Then we have Seuanez et al., 1977, from which no actual figures can be drawn, however it does highlight just how much variance is evident in previous results. Following that is Formenti et al., 1983, which – at least according to genomesize.com – does not support an 8% difference, but perhaps only a 2%-3% difference. And the last of these four is Pellicciari et al., 1990a, which supports a 7.5% difference, but also has quite a large standard deviation.
Contract this with the most recent study – Krishan et al., 2005 – which shows that the genomes are approximately the same size, and it’s difficult to see how Dr Tomkins can claim a majority in any sense of the word.
Without good reason to the contrary, I’m inclined to give more weight to more recent studies, and give less weight to older studies – particularly those with such enormous variance in their stated results.