BCH 447 lecture 3 notes
Natural selection is much more likely to dominate in a large population than
a small population because in a small population the chances are higher for a
gene to be lost and not fixed. Drift competes with selection.
All mutations have to survive the first initial stage. When they are first
provided to one individual. Regardless of small or large. But the chance of
something being lost through random genetic drift is the same in initiall
stages but decreases over time in a larger population.
Regardless of the population size, the beneficial mutation must reach a
certain threshold before it is immune to loss. This threshold is higher in a
smaller population. Natural selection is actually favoured in larger
populations and disfavoured in smaller ones. Genetic drift is more likely to
cause loss of a beneficial mutation in small populations and less likely to
do the same in larger ones. This does not mean that drift doesn't happen in
The reasoning why natural selection predominates in large populations is
because there is such a high concentration of mutation in large populations.
So the rate of fixation of genetic drift will be the same regardless of
So, small populations lose have few mutations and therefore few mutations are
Large populations have many mutations and many mutations are lost. But
because they have so many mutations the likelihood of one being fixed is
The number of mutations increases with a larger population size. And
therefore there are more mutations in large populations... As such there is a
larger likelihood that one gets fixed.
The reason we see a molecular clock for all organisms is because the rate of
substitution of amino acids is the same in both lineages independent of
For various lineages the rate of evolution is relatively constant over time.
We picked up the 1963 margoliash paper:
Side note, horizon thresholds.. It's a look back time. The further back we go
the scale of remembering shortness. We can't distinguish between 10000 and
1000 years ago. 1963 is well remembered by our proff. Lol
Back to 1963:
Homologus is defined as descent from a common ancestor. When proteins are
homologous it means that they are evolutionarily related. This is a strict
definition. Proteins with the same function may or may not be homologous...
An example of same function but non homologous is different actions in the
citric acid cycle are carried out by enzymes that are completely unrelated to
each other. These are examples of convergence into similar functions but non
The tricky part comes when we have intermediate commonalities. The cutoff for homology is. Homology has been used as a problem for evolution... We are
saying that these two proteins descend from a common ancestor, therefore
that's proof that they descend from a common ancestor. This is not the
argument science makes. We say, these proteins are really similar in
genetics... They are homologous, therefore it is likely that they descended
from a common source. It is not a problem. homology is a conclusion based on
Margoliash says... Homology is a function of similarity. And our cutoff for
homology is around 20%. Anything below that... It could've been chance. 20%
is the twilight zone.
Homology is an all or none situation. Things are either homologous or aren't
homologous. Nothing is 50% homologous. Things are 50% similar.
The data is similarity the conclusion is homology.
In areas where the heme group is attached to cytochrome C there are no
changes in amino acid sequences. All the proteins have the same amino acid
sequence. These guys are conserved because of function, we don't mutate them
because they are likely lethal.
If mutations are lethal we won't see them exist... Very important point.
The only plausible explanation for sequence homology is because the
individual genes must be related. They must descend from a common ancestor
and changed over time. This is why a tree can be formed.
Thus there is little doubt that the cytochrome C's are homologous to each
Margoliash and the amino acid sequencing of cytochrome c showed evidence for
relationships between organisms.
In 1963 there was no genetic code.
note, that in every position of a protein we will not find every possible
amino acid. But the reasoning behind it is not that the amino acid didn't
mutate to whatever it wants. It just means whatever had that mutation died.
This is why there is conservation even in mutation... Ex, important
hydrophobic or hydrophilic sectors tend to remain as such.
There are two possibilities for conservation of sequence. It's either that
only a certain subset of amino acids can replace a certain protein sector
because it would be detrimental. Or... There is biase in mutation... Such
that mutations prefer to mutate to common amino acids, ex... Luecine to
The truth is explained above the above paragraph.
Interestingly in contrast to the logic that any mutation can occur but some
will cause death and are therefore not seen. It seems that there are in non