Questions asked about the genetics lectures

These questions and answers are presented oldest first. So any questions asked as a result of recent lectures / exercises will be near the bottom of the list.

  1. Question I have some questions about some of the items on your webpages. In lecture 2 you say that after 6 or 7 divisions, some cell types have become irreversibly programmed to give rise to a subset of possible cell or tissue types. Is the nature of this 'programming' understood, ie does it involve DNA modification?

    Answer The process is only very partially understood. Given 'Dolly' my use of the term "irreversible" is perhaps wrong, the programming of a nucleus obviously can now be reversed sometimes (its still true of the cells however). The modification does not involve DNA modification. It is caused by the activation and inactivation of genes by the binding of control proteins. Although each of the activation / inactivation steps is in principle reversible, because of the complexity of the steps involved it is effectively impossible to backtrack. We now seem to have learned how to erase that programming by rendering cells quiescent and then by transplanting the nucleus into an egg. But the cells themselves remain irreversibly committed to a certain pathway of development.

  2. Question If so, is it known how the udder cell nucleus was 'reprogrammed' in the case of Dolly,

    Answer My knowledge of this is very little. I think that it is not known in detail - we don't know which genes were reprogrammed. We only think that everything was turned off.

  3. Question Could this in theory be done for any cell type?

    Answer I suppose so. So long as it has all its genes intact. It wouldn't work, I don't suppose, for a T or B cell and it certainly wouldn't work for an erythrocyte!

  4. Question Have only female cells been cloned?

    Answer I think so but I'm not sure. There was a recent paper in Nature on cloning mice.

    T. WAKAYAMA et al: Full-term development of mice from enucleated oocytes injected with cumulus cell nuclei. Nature 394, 369-374 (1998)

    in which cells derived from females appeared to do better than cells derived from males. However, they did not make strenuous efforts to obtain males and did in fact see one perfectly healthy appearing embryo at the 12 somite stage. So my impression is that there is no real barrier to cloning males.

  5. Question Why is it possible to cryopreserve early stage embryos but not eggs?

    Answer I don't know! maybe because the cells are smaller and therefore its easier to preserve their structure.

  6. Question And with regard to human 'cloning', if the male pronucleus of a fertilised egg was replaced with a female pronucleus from the same mother..in theory.. would the resulting daughter be a 'clone' of the mother or not as recombination has occured in the germ cells?

    Answer I'm glad you said "in theory"!

    I think not. The resulting daughter would not contain all of the mother's genetic information and would therefore not be a clone. However, the baby would not survive. Experiments with mice show that this develops abnormally giving very poor placental development. The converse - where you have two male pronuclei gives a very vigorous placenta and a poor embryo. It can even become malignant. This is part of the evidence for the existance of imprinting

  7. Question Are most polymorphic VNTRs/mini/microsatellite repeats confined to junk DNA or can they occur nearly anywhere, say in introns or promoters?

    Answer Yes to introns.  I can't think of any examples in promoters though.

  8. Question How likely is it that a given VNTR can expand between generations?

    Answer I'm not sure about minisatellites.  But microsatellites can mutate about once in every couple of hundred meioses.  When we look at the reference families which were used to construct the human genetic map we find mutations at about that frequency.

  9. Question In DNA fingerprinting, if there are loads (thousands?) of polymorphic markers how is it decided which ones to look at, and is there a minimum number of polymorphisms that should be looked at - or the more the better- and is this standardised between labs?

    Answer You do the ones which are known to work best - not all are equally easy to work with.  You do enough to get a significant result.  For forensic purposes there is some standardisation of choice of loci to avoid having to revalidate the system in every court case.

  10. Question In practice, how many Centimorgans apart are genes that can be picked up by linkage analysis, or does this just depend on the number of families that are being looked at?

    Answer It does depend on the number (and size) of the families that you have available.  If you had an infinite supply of families then you are limited only by the effort that you want to put into typing them.  In practice this means that you can probably find a marker within 5 cM without too much effort.

  11. Question Am I right in thinking that 'phase known' is the same as knowing if two markers/genes are in coupling/repulsion?

    Answer You are.

  12. Question In constructing a genetic map based on Drosophila progeny phenotypes, how do you know if the gene order is telomere to centromere or vice versa?

    Answer You don't.  That has to be worked out some other way.

  13. Question If a very high lod (Z) score is obtained for theta approching zero, for a polymorphism linked to a particular disease, is this evidence that the polymorphism is actually in the gene

    Answer Only that you can't separate the polymorphism from the disease causing mutation.  If you had chosen to work with a polymorphism within a gene which was a candidtae for some other reason you might be spurred to greater efforts by this result.

  14. Question Is it right to say that if z goes to minus infinity as theta approaches zero then the polymorphism cannot be in the gene?

    Answer Well, you've certainly found recombination between the site of the polymorphism and the site of the disease causing mutation.  So its odds on that they are some distance apart and not right next to each other.

  15. Question How is it that one third of cases of x-linked recessive lethal diseases are new mutations?

    Answer One third of the mutant X chromosomes are being removed by natural selection from the population each generation.  Therefore, because the disease remains at the same frequency from one generation to the next,  they have to be replaced.  Therefore the same number of new mutant chromosmes will be created.  And, because one third of the population's X chromosomes are in males, one third of the mutations will happen on chromosomes which end up in a male.  And so it follows that one third of the new mutations are removed immediately from the population by occuring directly in those X chromosomes which go to males.

  16. Question Can mutation rates be determined easily or just speculated on?

    Answer Yes and no!
    Its not easy but it can be done.  You can also make estimates based on the population frequency and Hardy Weinberg.  I'm not going to try here. Consult a textbook!

  17. Question Uniparental disomy:
    This is what I understand: both mutant alleles come from 1 parent, so the gamete from this parent must have had a non-disjunction at meiosis so that both copies went into one gamete.
    What I don't understand: how can a carrier parent have 2 mutant alleles in the first place???

    Answer The parent doesn't have two mutant alleles but transmits the one (s)he's got twice. i.e. it could be a failure of the second meiotic division for instance, or it could be that a monosomic zygote luckily duplicates the one chromosome which it has obtained or ... make up your own scenario!

  18. Question Mutation frequency of X linked recessive conditions:
    You mentioned in your lecture that since 1/3 of X chromosomes are in males, 1/3 of mutant chromosomes will therefore be in males and thus if the condition is lethal, 1/3 would be lost in each generation. if frequency of disease remains the same, that must mean that new mutations are giving rise to 1/3 of the mutations.
    What I don't understand: what about affected females (homozygous recessives)? Is this proportion insignificant to the above calculation?

    Answer Thoughtful point. It is insignificant though because the mutant allele is relatively rare compared to the non mutant. So the chance of a new mutation occuring on the other X chromosome of a female who was otherwise destined to be a carrier is very low. And how else are you going to get affected females? (The answer to that one is non random X inactivation).

  19. Question this is still on X linked recessive inheritance:
    if mutation frequency of such a gene is p, what is the frequency of affected individuals? How does the Hardy Weinberg equilibrium figure in THIS type of inheritance?

    Answer If the mutation frequncy is p then the disease frequency is 3p if the gene is lethal. Hardy Weinberg doesn't really apply because of the severe selection. If you were just asking what has HW to say about X linked conditions of minimal effect (say colour blindness for instance) then if f(A)=p and f(a)=q then the genotype frequncies of A and a males are p and q respectively. And the genotype frequencies of AA Aa and aa females are p2 2pq and q2 respectively as usual.

  20. Question I have a question about mutation:
    alu and L1 repeat elements - what are they? Are these already present in the genome or are they transmitted by viruses etc, and am I right in saying that their random insertion/excision can cause mutation?

    Answer alu and L1 elements are part of our normal genomic DNA. The typical alu element is about 300 bases long whereas L1 elements are much longer. The human genome contains as many as 500,000 alu elements randomly scattered throughout its sequence. There are also a lot of L1s (and many other different repetitive elements too). Both elemnts are capable of being transcribed and their transcripts can then reintegrate into the genome at a new site. (This does not happen very often but sometimes we can see its consequences. For instance there was once a new alu inserted at a particular site on the Y chromosome and there are now two sorts of human Y chromosome, those with the extra alu and those without.) The arival of an alu or L1 can cause a mutation but I can't think of an example off the top of my head.

  21. Question Also, why do 2 close alu elements make the chromosomes misalign at meiosis?

    Answer They don't force the chromosomes to misalign but they do make misalignment easier. Alu #1 can pair with alu #2 on the other chromosome instead of with its own homologue. This will have bad consequences if the misaligned alus then indulge in recombination. For instance, there is a recurrent mutation which deletes the promoter and first exon of the Adenosine Deaminase gene, ADA, which is caused by recombination between two alu elements, one upstream of the gene and one in the first intron.

  22. Question Can you answer some questions about past Royal Free exam papers?

    Answer Yes but this is a bit parochial so Royal Free students should click here.

  23. Question I am very confused about what germline mosaicism actually is. I have puzzled over the lecture notes and read the relevant pages in the textbook, and would really appreciate a simple explanation!

    Answer The gonads of a developing human embryo are a somatic structure which gets populated by a couple of million germ cells.  The germ cells go on to give rise either to oocytes or spermatogonia depending on the sex of the gonad concerned. You will hear about this in the lecture which Robin Lovell-Badge gives. Those millions of germ cells are derived from a much smaller number of progenitor cells. Suppose that a mutation occurs in one of the progenitor cells. Then the gonad will then become populated by two sorts of germ cells, normal ones and mutant ones.  In other words it will be a "germ cell mosaic".  The gonad's owner will be unaware of this.  The consequence however is that that person may have mutant offspring in unpredictable numbers - it all depends on the proportion of the germ cells which carry the mutation.  As a rule of thumb, we say that if a person has a child who is carrying a new mutation - and we can prove this by examining DNA from both parents and from the child - then the risk of having a second child with the same mutation is not extremely low as you might think, but is about 5%.  This is because the mutation event may have happened, not in the gamete which gave rise to the child, but much earlier in gonad development so that there are potentially many more mutant gametes waiting their chance.

    You might also look at exercise 3 in the CAL pedigree exercises which we did in Foundations of Health and Disease last year.

  24. Question  I was wondering if genetic mapping is a physical process and if so how it relates to FISH because I have found a few questions on relative distances between genes and I don't know how to interpret them.  Is genetic mapping to do with distances or order and what is it used for if we can find the same information using FISH.

    Answer  Genetic mapping is not a physical process.  It depends on the likelihood of a recombination event occurring at meiosis between the loci concerned.  On average, the likelihood of a recombination event is about 1% per million base pairs.  However, this can vary locally by a factor of maybe 10 fold (maybe more).  Genetic maps put loci in the correct order but the distances are "genetic" rather than physical.

    FISH can give you information about order and distance too.  But it is limited in two ways.  If loci are close to each other - say within 2 - 5 million basepairs -  then standard metaphase FISH will not resolve them.  More importantly, what if one of the loci is unknown and all you have is information that certain members of a family have inherited a disease (for instance as in the Indian eye anomaly story in the CAL).  What do you FISH with?  You have no probe, only a clinical phenotype.  In this circumstance all that you can do is map the disease by genetic (linkage) mapping.  When you know roughly where it maps you can examine the relevant region of the human genome for "candidate" genes.

Thanks to several students for the intelligent questions.