Science Questions

How many genes do humans share?

Tue, 9th Aug 2016

Listen Now    Download as mp3 from the show Do Fish Fart?


John Frederickson asked:

I know that humans generally share 99% of our genes. With 25,000 genes, that means we differ by only 250 genes. It seems to me that we cannot possibly differ by the same 250 genes since mutation, random assortment and crossing over are all random processes. Thus, my question is, how many genes does a random pair of humans actually share. Thanks.


We put this to Naked Scientist Kat Arney...crowd

Kat - And this is a great question, but itís sort of based on a misunderstanding. Because this is a phrase that we often hear that humans, we share 99% or 99.9% of our genes with each other. And, actually, if you took two random humans, there would be 4 million differences in the letters of our DNA. These chemicals called bases (theyíre like the letters of the alphabet of our DNA).

But the key thing is that the way Johnís phrased this question.  Itís like if you imagine our genome as a recipe book that makes all the recipes that our cells need, youíre imagining that we would have say 250 recipes, whole recipes that were different between us. Thatís not the case. Itís more like there is 0.1% of 4 million differences in single letters (kind of typos) scattered through the whole of this recipe book. And, obviously, between males and females (people who are genetically male and genetically female), thereís a whole chromosome difference.

So the two women in this studio, myself and Eleanor, we have two X chromosomes whereas the chaps here, Iím assuming, have and X and a Y chromosome. I havenít karyotyped everyone - I donít know for sure, but thatís my assumption.

So basically, itís not that we have whole genes that are different, thereís this kind of smattering of variations through the whole thing. However, some people do have changes, mutations, variations that do mean that whole genes, or even whole chunks of DNA are missing, or even whole bonus chunks. In Down Syndrome, youíve got a whole extra copy of one of  the chromosomes. So there are lots of variations between us but itís not like saying this whole gene is there, this whole gene is different.

Chris - Is it a bit like - you used a recipe analogy Kat - so youíre making a cake and it says youíve got to have the flour, and the eggs and the margarine, and the raisins, and so on. And rather than getting your raisins from that shop, youíve gone and got a sort of different type of raisin. Theyíre still raisins but theyíre slightly different raisins and, therefore, the recipe you cook up will make a slightly different cake but itís effectively still a fruit sponge?

Kat - Yes. Or it could be raisins versus sultanas, or oranges versus lemons. And, actually, just in our own genes we have a lot of variations. We make, if you think of it in terms of recipes, we make several hundred thousand different recipes called ďproteinsĒ that make ourselves function, keep us functioning healthily, but we only have about 20-25,000 genes. So thereís a lot of switching in and switching out anyway, and these tiny, tiny variations scattered between them make us all unique and different.


Subscribe Free

Related Content


Make a comment

Most (over 90%) of human DNA sequence is "junk" non-coding ...

Mutations in junk DNA may have no consequence , so those regions can have more variability between individuals than in the regions with "essential genes" ... ].

The "junk" DNA is used in genetic fingerprinting ... RD, Tue, 5th Aug 2014

This reply doesn't answer the question. The 25,000 genes reference all coding genes identified by the Human Genome Project.  The question is: How many genes do a random pair of individuals actually share, given the random events involved in creating each person's genome.  The On-Line Mendelian Inheritance in Man ( currently lists 3251 genes with phenotype-causing mutations.  This is, roughly, 13% of our genome, not the 0.1% cited by the Human Genome Project. JackDarwin, Wed, 6th Aug 2014

A Clarification: Two people can share the same protein-coding gene, but have different variants of the same gene. Is the OP asking about genes or gene variants?

These variants can be Single-Nucleotide Polymorphisms (SNPs), or larger changes like chopping out part of the gene, or duplicating part of the gene - or even deleting the whole gene.

Many SNPs do not affect the function of a protein - but if it occurs in the critical binding area of an enzyme, it could block the normal function of that enzyme, or produce unintended actions.

Even if one copy of the gene is damaged, we have a backup copy on the other chromosome. Recessive genetic diseases only take effect if you inherit two copies of the affected gene (apart from males, who don't have a backup copy of the X & Y chromosomes).

Since there are 64 DNA codons, but just over 20 amino acids used in proteins, there are multiple codons which will produce the same amino acid and protein. Such a genetic mutation (change to the genome) has no impact on the protein (proteome).

On the other hand, some mutations in non-coding areas (eg a regulatory region) can have as big an impact on the proteome as deleting an entire gene.

The probability of passing on a mutation in the population will vary from gene to gene - a non-critical gene could be lost, with nobody noticing, while critical genes or dominant mutations can have a life-threatening impact.

This wide range of potential impacts allows people to generate and quote a wide range of statistics on genetic variability to prove whatever they want. evan_au, Wed, 6th Aug 2014

See the whole discussion | Make a comment

Not working please enable javascript
Powered by UKfast
Genetics Society