'Crossing Over'

334
1
Hello.
I have a little question about 'crossing over'.
Sorry for the scientifical terms, i was unable to find all of them, so i guess this will be hard to understand (i would appreciate if you correct me in spelling, and give me the scientifical terms that i missed).
I understand tha during crossing over, part of the non-identical chromatides forming a tetrad are exchanged.
I also know that the chromatides are made of a DNA molecule (and ... proteins ?), which are made of neucluotides, and that a gene is a group of codons, being 3 neucuotides.
So, in conclusion, during 'Crossing over', some neucluotides are exchanged.
Now, my question is, is it possible that not a whole gene gets exchanged between the two chromatides (iow, half a gene, or quarter a gene). If so, wouldn't this cause a lot of problems ?

Thanks (if you see the question is not clear enough, please ask me to clear it :smile: ).
 

iansmith

Staff Emeritus
Science Advisor
Gold Member
1,345
2
Originally posted by STAii

Now, my question is, is it possible that not a whole gene gets exchanged between the two chromatides (iow, half a gene, or quarter a gene). If so, wouldn't this cause a lot of problems ?
Yes this can happen because recombination occurs at homologus sequence sites. It does not create problem if both the first half is complementary to the second half. Meaning that the gene will not become non-functional. It can cause problem when recombiantion occurs at not some homologous sites (i.e. 2 similar sequence but not complementary to each other). Then a gene and also a chromosome can become non-functional.

Also recombination also used to introduce mutation in genetic enginerring. And recombination can also occurs at specific sites and site specific recombination requires special enzyme, which are called recombinase.

some definition

Chromatid: a portion or strand of chromatin

Chromatin: The chromosome as it appears in its condensed state, composed of DNA and associated proteins (mainly histones).
 

Monique

Staff Emeritus
Science Advisor
Gold Member
4,069
62
Hi Staii, crossing over occurs during meiosis, the forming of gametes. At this time the two same chromosome pairs are alligned (say twice chr. 19) to be seperated from eachother to form a haploid cell.

Here is a good image of what happens: http://www.uni-kiel.de/medinfo/mitarbeiter/krawczak/folien/vorlesung9/img004.jpg [Broken]

So yes, a chromatid is a single strand of a DNA molecule, the upper portion of the image is the tetrad. You should understand this is a dynamic structure! The tetrad is not fixed at a point in the strand, but can move up and down like a zipper.

So, on the DNA there are genes. A gene consists of coding and non-coding regions. The coding regions have a three-letter code: codons, which will become a certain aminoacid during translation.

But then you go wrong, nucleotides are not exchanged, an entire arm of the chromosome is exchanged! So this is something which happens at a single point, as is clear in the image.

So it could be that half the gene is paternal and the other half of the gene is maternal, yes. But then you ask if it is possible to get a quarter gene. This is actually a very good question.

As you can see, a single cross over (recombination event) results in half blue, half red. Now, during meiosis about 40 of these cross-overs occur. So an alternating pattern of blue/red is possible. The distances are restricted though!!

I forgot what it is called, it has a special name. Basically, if a cross-over has happened somewhere on the genome, it is highly unlikely - if not impossible - for a second crossover to occur in the same fecinity.

Problems don't occur, as long as the cross over is symmetric.
 
Last edited by a moderator:

Monique

Staff Emeritus
Science Advisor
Gold Member
4,069
62
Ian has beaten me :)

Just to make it clear, during meiosis there is a process that inhibits recombination happening near the same spot twice.

Genetic engineering though IS possible, where such homologous recombination exchanges a very specific piece of DNA.. but there are also length restriction, I am not aware how long since I never actually did the experiment :)
 
334
1
But then you go wrong, nucleotides are not exchanged, an entire arm of the chromosome is exchanged!
You are right, i mis-expressed.
So it could be that half the gene is paternal and the other half of the gene is maternal
What do you mean ?

Ok, back the subject.
I am not suggesting that the proccess happens twice.
Let's go back to your drawing.
In your drawing, the genes that the drawing was focusing on where not in the exact point of the chiasma (A,a,B,b are all not in the point of chiasma).
Now, suppose that A (for example) was at the exact point of chiasma, a certain percentage of it was on the right side of the chiasma, and another part was on the left part of the chiasma.
This way, only part of the gene A will be transported from one chromatid to another.
This is what i am talking about, is this possible ?
 

Monique

Staff Emeritus
Science Advisor
Gold Member
4,069
62
I remember now! It is called cross-over interference.
 

Monique

Staff Emeritus
Science Advisor
Gold Member
4,069
62
Yes, that is what I meant :)

You have got two Chr. 19, one from your father, one came from your mother. These two allign. Cross over occurs in the middle of the gene. Now that gene on the chromosome consists part of the paternal and part of the maternal DNA (ofcourse the DNA is all yours, but it makes the discussion clearer as to the origins of the DNA).
 
334
1
Umm .. weird.
Now, the gene is a sequence of neucluotides.
If two different sequences (ie, two genes) where exchanged from the middle, how do we know that the result actually has any meaning ?
Suppose (for example) that a gene is :
AAATTGGCC <-- from father
GCCTAGCCT <-- from mother
.. Crossing over after the forth neucluotide ...
AAATAGCCT <-- New 1
GCCTTGGCC <-- New 2
The old sequence had a certain meaning (it was a working gene), but how do we know that the new gene has any meaning ?
 

Monique

Staff Emeritus
Science Advisor
Gold Member
4,069
62
This image is actually a better representation of the chromosomes:
http://www.mds.qmw.ac.uk/statgen/dcurtis/lectures/meiosis.gif [Broken]

I am not sure how to interpret your sequence, staii.. let me say this: we share 98% of our DNA with the chimps.. how much is the same between humans? 99.9xx%

Our genes are the same, the alleles make us different from eachother. So it doesn't matter if the first half of the gene has a maternal origin and the other half a paternal origin. As long as the cross over is symmetric and the ORF (open reading frame of the codons) is not changed.
 
Last edited by a moderator:
334
1
From what you say, it seems like any random neucluotides sequence will be meaningfull (and functional) ?
Anyway ...
Can you explain this :
Open reading frame of the codons
Thanks for all the help

(BTW, this seems like a chat more than being on forums)
 

iansmith

Staff Emeritus
Science Advisor
Gold Member
1,345
2
Originally posted by STAii

Suppose (for example) that a gene is :
AAATTGGCC <-- from father
GCCTAGCCT <-- from mother
.. Crossing over after the forth neucluotide ...
AAATAGCCT <-- New 1
GCCTTGGCC <-- New 2
The old sequence had a certain meaning (it was a working gene), but how do we know that the new gene has any meaning ?
Staii, your example will not probaly happen because there is not enough homology between both sequence.

Here a better example

AAATATATATAGTGCTATATATTTTTATTTATTATGGTG <-----Father

AAATATATATAGTGCTATATCTTCTTGTTTATTATGGTG <-----Mother

Cross over befor the bold

AAATATATATAGTGCTATATCTTCTTGTTTATTATGGTG <-----New 1

AAATATATATAGTGCTATATATTTTTATTTATTATGGTG <-----New 2

But lets translate the sequence to aminio acid

Lys-Tyr-Ile-Ter-Cys-Tyr-Ile-Phe-Leu-Phe-Ile-Met-Val <-----New 1

Lys-Tyr-Ile-Ter-Cys-Tyr-Ile-Phe-Leu-Phe-Ile-Met-Val <-----New 2

Both a.a sequence are identical. As we said before it will not matter where the cross over occurs unless the ORF is change, the a.a. sequence is change significantly or the cross over occurs at a similar site but not complementary

Example

#19 AAATATATATAGTGCTATATATTTTTATTTATTATGGTG <-----Father

#14 AAATATATATAGTGCTATGGTAGGTAGTAGTAGTTTTATTATGGTG <-----Mother

Double Cross over befor and after the bold

AAATATATATAGTGCTATGGTAGGTAGTAGTAGTTTTATTATGGTG <-----New 1

AAATATATATAGTGCTATATATTTTTATTTATTATGGTG <-----New 2
 

Monique

Staff Emeritus
Science Advisor
Gold Member
4,069
62
hehe, you know we will charge you for this :wink:

With ORF the way the codons are read from the site of initiation. It goes in pairs of three right? The following is an ORF of: ACGGTCCCGTTC

ACG GTC CCG TTC in this way it will be translated the wrong ORF would be caused by the insertion of a base for instance:

ACG aGT CCC GTT C

From what you say, it seems like any random neucluotides sequence will be meaningfull (and functional) ?
Anyway ...
Why do you say that? The SNP consortium has found 1,255,326 SNPs, Orchid biosciences an additional 10,000.

Btw, it is nucleotide not neucluotide
 
Last edited:

Monique

Staff Emeritus
Science Advisor
Gold Member
4,069
62
nice one Ian, but just remember that in vivo things don't happen like that due to cross-over interference.
 

iansmith

Staff Emeritus
Science Advisor
Gold Member
1,345
2
Originally posted by Monique
nice one Ian, but just remember that in vivo things don't happen like that due to cross-over interference.
Yeah but I think we will lose him if it gets too complicated with all the exemption. I tried to keep thing simplified:wink:
 
334
1
Originally posted by iansmith

But lets translate the sequence to aminio acid

Lys-Tyr-Ile-Ter-Cys-Tyr-Ile-Phe-Leu-Phe-Ile-Met-Val <-----New 1

Lys-Tyr-Ile-Ter-Cys-Tyr-Ile-Phe-Leu-Phe-Ile-Met-Val <-----New 2

Both a.a sequence are identical. As we said before it will not matter where the cross over occurs unless the ORF is change, the a.a. sequence is change significantly or the cross over occurs at a similar site but not complementary
Isn't this a coincidence? I noticed that the two original genes are also translated to the same amino acid sequence.

EDIT :
Btw, it is nucleotide not neucluotide
Thanks, i highly appreciate this (and everything else of course), but getting the right terms is always hard when u study in a different language.
 
Last edited:

iansmith

Staff Emeritus
Science Advisor
Gold Member
1,345
2
Originally posted by STAii
Isn't this a coincidence? I noticed that the two original genes are also translated to the same amino acid sequence.
This is not a coincidence, I did it on purpose, I change the nucleotides sequence but I kept the same amino acid sequence. It was to show you that the DNA sequence is not that important if the amino acid sequence is the same or relatively the same and the protein is functional.
 
334
1
Yes, but in nature, i don't think that the amino acid sequence will always stay the same (if so, there will be no benefit of 'crossing over')
So, if i am right, the amino acid sequence will actually change, when it changes, what makes the new sequence meaningfull (ie, it will still act like the gene that was in this place before crossing over).

BTW, if you feel you are putting too much effort into this, forget about it.

Thanks.
 

iansmith

Staff Emeritus
Science Advisor
Gold Member
1,345
2
Originally posted by STAii
Yes, but in nature, i don't think that the amino acid sequence will always stay the same (if so, there will be no benefit of 'crossing over') So, if i am right, the amino acid sequence will actually change, when it changes, what makes the new sequence meaningfull (ie, it will still act like the gene that was in this place before crossing over).
Most of the genes do have a amino acid sequence that are identical or similar (meaning that a a.a from a certain group [+ charge, - charege, polar, non-polar and aromatic] is change by another a.a from the same group.

Some genes do have different a.a sequence. For example, a +ve charge a.a is change by a -ve charge a.a. This change migth not affect the function of the protein. In the other hand, the protein could become non-functional or functional under certain conditional (refer to as conditional mutant). Other mutation will introduce a stop codon and the protein becomes non-functional.

Most genes from individuals are not that different but some genes show a lot of different phenotype, eye and skin color for example. Crossing over can be beneficial for individuals that carries dangerous mutation. Crossing over also allow the exchange of gene and a mixture of genes in the population.
 

Monique

Staff Emeritus
Science Advisor
Gold Member
4,069
62
Basically you have got two genes, paternal and maternal. These two genes can be completely homologous, than both parents will have blue eyes (hypothetically). Or, the genes have a small difference in the first exon, two alleles. Now the one gene codes for blue eyes and the other for green eyes (hypothetically). Imagine crossing over occurs in the middle of the gene. Since only the first exon is different, and the rest is the same.. only switching of the genotype occurs from paternal to maternal chromosome.

You must understand that genes are highly conserved sequences, there really isn't that much variation in them to cause massive structural changes when they are interchanged.

I have sequenced genes of unrelated individuals, they are completely the same!

Crossing-over DOES have an evolutionary advantage though, a genome is reshuffled, creating more genomic diversity and less inbreeding, simply stated :)
 

Monique

Staff Emeritus
Science Advisor
Gold Member
4,069
62
It is also important to keep in mind that only about 2% of the genome is coding, less than 1% exons! (heard that in a lecture, not sure where to get the data from)
 

The Physics Forums Way

We Value Quality
• Topics based on mainstream science
• Proper English grammar and spelling
We Value Civility
• Positive and compassionate attitudes
• Patience while debating
We Value Productivity
• Disciplined to remain on-topic
• Recognition of own weaknesses
• Solo and co-op problem solving
Top