Scribd
Upload
27 views16 pages

Evolutionary Change of Amino Acid Sequences

Evolutionary changes to amino acid sequences can provide useful information about long-term gene and species evolution. Mathematical models can be used to analyze the rate of amino acid substitutions over time and between lineages. The Poisson correction model assumes amino acid changes occur at a constant rate and can be used to estimate the proportion of differences and rate of evolution between sequences. Bootstrapping methods allow estimating variances and covariances of distance measures without assumptions about independent evolution. Amino acid substitution matrices reflect differing substitution rates between chemically similar amino acids. Mutation rates provide an upper limit to actual substitution rates after accounting for genetic drift and selection.

Uploaded by

princez_shalini
Copyright
© Attribution Non-Commercial (BY-NC)
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PPT, PDF, TXT or read online on Scribd
27 views16 pages

Evolutionary Change of Amino Acid Sequences

Evolutionary changes to amino acid sequences can provide useful information about long-term gene and species evolution. Mathematical models can be used to analyze the rate of amino acid substitutions over time and between lineages. The Poisson correction model assumes amino acid changes occur at a constant rate and can be used to estimate the proportion of differences and rate of evolution between sequences. Bootstrapping methods allow estimating variances and covariances of distance measures without assumptions about independent evolution. Amino acid substitution matrices reflect differing substitution rates between chemically similar amino acids. Mutation rates provide an upper limit to actual substitution rates after accounting for genetic drift and selection.

Uploaded by

princez_shalini
Copyright
© Attribution Non-Commercial (BY-NC)
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PPT, PDF, TXT or read online on Scribd
You are on page 1/ 16

Evolutionary change of Amino

Acid Sequences
• Before DNA, study on mol evo were
conducted using a.a seq data.
• DNA is much more simpler but a.a are more
conserved than DNA- provide useful
information on long-term evolution of genes
and species.
• But, mathematical model for a.a is much more
simpler.
Amino acid differences and proportion
of different amino acids

• Compare the the number of amino


acid differences
Eg: number of a.a that differ from a
human hemoglobin
Model
• assume that amino acid changes occur at a
rate λ per unit time
• – two independently evolving lineages,
initially
• identical in sequence
– at time t there are a proportion Dt differences
Poisson correction (PC) model
• assume that amino acid changes occur at a rate λ per unit time
– two independently evolving lineages, initially identical in
sequence
– at time t there are a proportion Dt differences
– at time t + 1
• D(t+1) = (1 - Dt) (2 λ) + Dt
• assume that amino acid changes occur at a rate λ per unit time
• D(t+1) = (1 - Dt) (2l) + Dt
• Dt = 1 - e -2 λ t

• expected proportion of difference, Ka


• Ka = 2 λ t or Ka = -ln ( 1 - D)
• assume that amino acid changes occur at a
rate λ per unit time
• total number of amino acid substitutions
• per site (expected proportion of differences)
–Ka = 2 λ t or Ka = -ln ( 1 – D )
–and rate of evolution is given by:
• – λ = Ka / 2t
• D = 0.56
• Ka = -ln (1 - 0.56 ) = 0.82
• λ = 0.82 / (2 x 80,000,000) = 5.1 x 10-9
• amino acid replacements per year
Bootstrap variances and covarians
• Bootstrap method is convenient to use to
compute the variances and covarians of various
distance measures.
• Bootstrap method requires no assumption-
evolve independently.
• Eg: random sample of 3 a.a seq with n a.a sites is
generated from the original data set.
• - the random sample is produced by sampling a.a
with replacement using pseudorandom numbers.
• X11, X12, X12, X14, X15...., X1n-1
X21, X22,X22, X24, X25,...., X2n-1
X31, X32, X32, X34, X35,...., X3n-1

This is called a bootstrap resampled data set,


and once this random samples is obtained, an
estimate of distance is computed.
• Adv of bootstrap
-capable of computing variances and covariances even
no mathematical formulas are available
-gives better estimates of the variances and covariances
than approximate analytical formulas (n>100).
• If the original sample is small and biased, this
biased cannot be removed by the bootstrap. In
this case, analytical formula give more accurate
variances and covarians than bootstrap.
Amino acids substitution matrix
• Substitution occur more often btw aa that are
similar in biochem properties( polarity,
volume).
• Some aa are rarely change. Eg cys, gly and
tryp.
• Theoritically, different proteins are expected
to have different substitution matrix for each
group of protein as more data on amino acid
sequences become available.
• In recent year, the a.a substitution matrix has
been used in the reconstruction of
phylogenetic trees by the max likelihood
method and in the inference of a.a seq of
ancestral proteins.
Mutation rate and substitution
rate
• We consider aa substitutions as though every
mutational change of aa were incorporated into
the seq under consideration. In practice, this is
not true
• Any mutation may disappear from the population
by chance or by purifying selecetion.
• When mutation spreads through entire
population, this is called fixation.
• Once mutation is fixed, every individuals of the sp
has the same mutation.
• When compare aa from 2 dif sp, we look at the
aa changes.
• For aa seq data, we usually consider the
substitution rate per amino acid site per year. If
we define the mutation rate as the rate (µ) per aa
site per year, the rule apply to neutral mutation.
Therefore, the rate of aa substitution per site per
year (r) is equal to mutation rate µ. That is

r= 2Nµ X 1/2N= µ
• Eg: n=100 aa, assume that each aa mutates to
other aa with the rate of 10-9 a.a per year/ 10-7
per polypeptide per year.
- If the effective size of population is 105, the total
number of mutation is
2 X 105 X 10-7=0.02 per year.

However, since they are fixed with the probability


of 1/(2 X 105), the rate of aa substitution per site
is r= 0.02/ (2 X 105 X 100)=10-9, which is equal to
the mutation rate per site per year.

You might also like