Researching the Genetics of Music, on OMIM

27 August, 2008
Looking for "music" in the Online Mendelian Inheritance in Man website.


OMIM is the Online Mendelian Inheritance in Man website. To quote the website's own home page, it contains "information on all known mendelian disorders and over 12,000 genes".

This number can be compared with the estimate of 20,000 to 25,000 protein-encoding genes in the human genome, i.e. about half. (At least it can be compared if we assume that most of those 12,000 are protein-encoding genes.)

The Scientific Investigation of Music

If there existed some clear non-human analogue of music, then the normal investigative methods of neuroscience could be expected to make steady progress in determining what music is, or what music perception represents, and how music is perceived and processed in the brain.

But, it seems very likely that the making and appreciation of music is a very specifically human behaviour, and this severely limits the directness of methods that can be used to investigate it. For example, fMRI is commonly used to observe "brain activity" of subjects performing various cognitive or behavioural tasks, but the picture it provides is a very fuzzy one, and the linear resolution is almost a hundred times less than what is required to measure and observe the activities of individual neurons and individual synapses (i.e. a typical fMRI "voxel" contains hundreds of thousands of neurons).

Finding Music in the Human Genome

Assuming that music, or at least music perception, represents a specific human adaptation, or a side-effect of such an adaptation, we would expect to see some kind of evidence on that adaptation in the human genome.

With the advent of modern sequencing technologies, the genomes of the human species, of different human individuals, and of other animal species, can all be investigated and described down to the very individual molecules that define them.

In other words, assuming that there is such a thing as a "musical" gene, i.e. a gene whose evolution caused the human species to become "musical", for whatever reason, that gene has already been described and sequenced, and can be viewed by anyone with access to the Internet.

The only problem, of course, is knowing which gene it is. Indeed, if there is one "musical gene", there is probably more than one. But where are they, and how do we find them?

To make good progress, in the absence of direct access to those brain processes involved in music perception, we need to formulate plausible hypotheses, and use these to guide our search. For example, my Musical Astrocyte Theory suggests that music perception may involve specific signalling between neurons that respond to the specific aspects of music and nearby astrocytes. Following this hypothesis, one could attempt to identify particular genes involved in neuron-astrocyte signalling, and look for genes where there is evidence of recent human-accelerated evolution in contrast to little or no evolution in other closely related species.

Search for "Music" in OMIM

However, there is one even simpler approach we can try first, and that is to search for the word "music" in the OMIM database.

As of today (27 August, 2008), this search gives the following seven results:

  1. Musical Perfect Pitch
  2. Williams-Beuren Syndrome
  3. Autosomal Dominant Lateral Temporal Lobe Epilepsy
  4. Tune Deafness
  5. Zinc Finger and BTB Domain-Containing Protein 33
  6. Osteopetrosis
  7. Kabuki Syndrome

Incidental Hits

Of these, the Zinc Finder gene is called "Kaiso" because someone was playing Kaiso music when they were analysing the gene in their laboratory, and one recorded sufferer of Osteopetrosis survived long enough to major in music at University.


In the case of the Epilepsy entry, you have to go the Clinical Synopsis to find the reference to music as one of the many complex auditory perceptions that may occur as part of a seizure

Musical "Ability" and "Disability"

The "Perfect Pitch" and "Tune Deafness" entries discuss two general aspects of musical "ability" and "disability" respectively. I have discussed both of these issues previously on my blog, here, here, and here.

So far the scientific literature on both of these subjects has very limited relevance to understanding of music perception, because:

Two Syndromes

Which leaves us with the two "syndromes". Williams syndrome and Kabuki syndrome are distinct syndromes, but they both involve some form of mental retardation, and they both include unusual degrees of interest in music.

Of the two, Williams syndrome is the more famous, and has been discussed by Oliver Sacks in his books and on a television program. Those affected by Williams syndrome are very sociable, although their social "skills" tend towards the positive rather than the negative, to the extent that pragmatically they are over-willing to socialise with whoever they meet.

Genetically, neither syndrome provides very specific information related to music perception.

In the case of Williams syndrome, a large (i.e. about 500kb) area in chromosome 7 is deleted. This area contains at least three known protein-encode genes, and probably more.

The genetics of Kabuki syndrome is somewhat more mysterious, because although genetic anomalies such as deletions and duplications have been detected in various patients, and inheritance of the condition has been observed, no single consistent genetic cause has been discovered. (Even if somehow there are several different Kabuki syndromes, one would want to know why they all showed similar symptoms.)

Music and Intelligence

Even though they fall short of giving us definite clues about the nature of music, the two syndromes come closer to this than anything else so far known about musical genetics.

The human brain is a complex thing, and genetic disorders which affect intelligence may do so either because they interfere with the brain's current operation, or because they interfere with its development. It's like throwing a spanner into the car engine before it comes out of the factory. Does the car go badly because the spanner stops the engine working, or because the presence of the spanner prevented the engine being built properly in the first place? The human brain is far more complex than a car engine, but at the same time, given its biological nature, it is much more robust and self-healing.

It is a general observation that interest in music is not particularly correlated with intelligence. People of high intelligence are no more or less interested in music than people of lower intelligence. People of "normal" intelligence can be amusic, and people of much less than "normal" intelligence remain interested in music.

This may indicate that music perception operates by means somewhat distinct from whatever constitutes "intelligence", or it may just be that "intelligence" is a much more fragile thing.

(This does raise an additional question, which is whether people with mental disabilities, be it Williams, Kabuki or anything else, are measurably different in their musical tastes from the rest of us.)

Locality and Globality

One of the specific claims of my own super-stimulus theory is that the perception of music is based on the summation of localised properties of activity in the cortical areas which respond to aspects of music. So one feature of disorders which interfere with normal intelligence may be that they disrupt the global structure of the brain, but leave its local structure intact, and as a result the perception of music is relatively unaffected.

Furthermore, one extension of my theory suggests that the final "calculation" of musicality is performed by astrocytes, with the "result" of this calculation transmitted non-synaptically back into the neural circuits. Disorders that interfere significantly with neuronal computation and synaptic connectivity may have less effect on this non-neuronal computation and non-synaptic transmission of information.


At the end of this analysis there isn't that much progress made. Perhaps the more critical "music genes" are yet to be found in the second 12,000 genes not yet included in OMIM.

They may even be found in the non-protein encoding portions of the genome. However, having considered this possibility, if music perception does involve mechanisms significantly different to normal neuronal computation (as suggested by my theory), this is very likely to involve specific protein-based adaptations.

In the mean-time, we can keep searching for "music", to see if anything turns up.

(End-note: you can also search via Google, e.g. site: music. This makes it easier to see where the word "music" appears, but unfortunately not all the relevant pages seem to be indexed by Google.)