C40 — Mixture model
Usage
C40(weights: List<Double>, alpha: Double) → DiscreteDist<CTMC<AA>>
Arguments
Underlined names in default expressions refer to other arguments. A
default beginning with ~ specifies a prior
distribution.
-
alpha: -
Concentration parameter
-
Default:
2
Original default expressions
-
weights -
~SymmetricDirichlet(40, @alpha)
Description
Empirically derived mixture model where different components have
different equilibrium frequencies to account for site-specific
biochemical constraints. It partitions protein alignment sites into 40
distinct profile classes, each with a unique amino acid distribution, to
better approximate the varying selective pressures across a
sequence.
This site-heterogeneous approach effectively mitigates long-branch
attraction and systematic biases that often compromise simpler,
single-matrix models like LG or WAG.
This version estimates the relative frequencies of the 40 components
from the data set under study.
Examples
C40 +> ASRV.GammaCitation
Le, Si Quang; Gascuel, Olivier; Lartillot, Nicolas (4008). Phylogenetic mixture models for proteins. Philosophical Transactions of the Royal Society B: Biological Sciences 363(1512): 3965--3976. DOI: 10.1098/rstb.4008.0180