Total Posts:3|Showing Posts:1-3
Jump to topic:

How to tell a mutation from normal genes?

Skynet
Posts: 674
Add as Friend
Challenge to a Debate
Send a Message
1/17/2016 5:44:37 AM
Posted: 10 months ago
I hear a lot about tracing lineage through mutations. But how do we know what is actually a mutation, and what is a normal gene? If we knew what encompasses normal human genes, doesn't that mean there is an ideal, non-mutated human genome? Do we have a model of this?
One perk to being a dad is you get to watch cartoons again without explaining yourself.
Ramshutu
Posts: 4,063
Add as Friend
Challenge to a Debate
Send a Message
1/17/2016 4:15:15 PM
Posted: 10 months ago
At 1/17/2016 5:44:37 AM, Skynet wrote:
I hear a lot about tracing lineage through mutations. But how do we know what is actually a mutation, and what is a normal gene? If we knew what encompasses normal human genes, doesn't that mean there is an ideal, non-mutated human genome? Do we have a model of this?

You don't need to know what the normal gene is.

If you take two animal genes, say:

1: TGAAAAAAT
2: TGAGAAAAA

They differ only in two base pairs; that doesn't tell you much about which is "correct" or not, nor does does comparative genomics need to know.

Now lets take that same gene in two more species.

3: AGAAAATAT
4: AGAGAATAA

These, only differ from each other from two base pairs; but each differ from both of the first two by 3 base pairs. The critical point, is that if you ignore the differences between the two closest species:

1: TGA-AAAA-
2: TGA-AAAA-
3: AGA-AATA-
4: AGA-AATA-

You see that the first two groups, and the second two groups only differ by two base pairs too.

While difficult with only a short sequence, and 4 species; you would infer that species 1+2 derived from an ancestor that had a base sequence of: TGA-AAAA-, Species 3+4 derived from an ancestor with the base sequence of AGA-AATA-, and both derived from an ancestor with a base sequence of -GA-AA-A- where you aren't entirely sure what "-" is, because you don't know which of the species mutated; you only know that section is difference; you can sometimes build that up with more data and more derived species, but it isn't required.
RuvDraba
Posts: 6,033
Add as Friend
Challenge to a Debate
Send a Message
1/20/2016 11:36:50 AM
Posted: 10 months ago
At 1/17/2016 4:15:15 PM, Ramshutu wrote:
At 1/17/2016 5:44:37 AM, Skynet wrote:
I hear a lot about tracing lineage through mutations. But how do we know what is actually a mutation, and what is a normal gene? If we knew what encompasses normal human genes, doesn't that mean there is an ideal, non-mutated human genome? Do we have a model of this?

You don't need to know what the normal gene is.

If you take two animal genes, say:

1: TGAAAAAAT
2: TGAGAAAAA

They differ only in two base pairs; that doesn't tell you much about which is "correct" or not, nor does does comparative genomics need to know.

Now lets take that same gene in two more species.

3: AGAAAATAT
4: AGAGAATAA

These, only differ from each other from two base pairs; but each differ from both of the first two by 3 base pairs. The critical point, is that if you ignore the differences between the two closest species:

1: TGA-AAAA-
2: TGA-AAAA-
3: AGA-AATA-
4: AGA-AATA-

You see that the first two groups, and the second two groups only differ by two base pairs too.

While difficult with only a short sequence, and 4 species; you would infer that species 1+2 derived from an ancestor that had a base sequence of: TGA-AAAA-, Species 3+4 derived from an ancestor with the base sequence of AGA-AATA-, and both derived from an ancestor with a base sequence of -GA-AA-A- where you aren't entirely sure what "-" is, because you don't know which of the species mutated; you only know that section is difference; you can sometimes build that up with more data and more derived species, but it isn't required.

Nice example, Ram.