GENE TREES Abhita Chugh. Phylogenetic tree Evolutionary tree showing the relationship among various entities that are believed to have a common ancestor.

GENE TREES

Abhita Chugh

Phylogenetic tree

Evolutionary tree showing the relationship among various entities that are believed to have a common ancestor

Species tree

• A phylogenetic tree showing the relationship among various species that are believed to have a common ancestor

Species tree

Shows the evolutionary history of a set of species

Speciation Nodes

Gene tree

• A phylogenetic tree that depicts how a single gene has evolved in a group of related species

• For this talk, evolve = duplication or loss

• Can be constructed over the topology of a species tree

Gene tree

Shows the evolutionary history of a single gene

Speciation Nodes

Duplicationnodes

Some definitions: Homologs

• Homolog: A gene related to a second gene by descent from a common ancestral DNA sequence

• Two types:

(i) Orthologs

(ii) Paralogs

Orthologs

• Genes in different species that evolved from a common ancestral gene by speciation - Retain the same function

Primates

Human Chimp

Speciation

Paralogs

• Genes related by duplication within a genome

• Evolve new functions

Primates

Chimp HumanRat

Rodents

Mouse

Why are Gene Trees interesting?

• Determine the evolutionary history of a gene family

• Infer gene duplications and losses

• Estimate bounds on times these events occurred

• Determine whether a given pair of homologs is orthologous or paralogous

Gene tree can be constructed over a species tree topology

PRIMATES INTELLIGENCE

No, seriously ..

Gene Tree Reconstruction

• Problem: Given a set of sequences from a gene family, find the tree that best explains the data

• 2 models:– Micro-evolutionary: considers sequence

evolution only– Macro-evolutionary: considers duplication

and losses only; useful but rarely used

Macro-evolutionary Problem

Macro-evolutionary Problem

Reconstruction algorithm

• Only macroevolutionary events are considered

• i – number of gene copies a node inherits from its parent

• j – number of gene copies a node sends to its children

• Range from 1 to m, where m is the maximum multiplicity of the gene in any species

Reconstruction algorithm

• The entering number of genes in root should be one• For each node, v, the dynamic program calculates the

minimum D/L Score of the subtree rooted at v, for all possible values of i and j

Step 1: Annotates minimum cost tables for all nodes

• cost [ i, j ] = cost at a node if it inherits i genes and sends j genes

• cost [ i ] = minimum cost at a node if it inherits i genes

= minimum { cost [ i, j ] }, for all j

cost[1] = 1cost[2] = 0

cost[1] = 1cost[2] = 0

cost[1] = 0cost[2] = 1

Cost of an internal node = cost of duplication/loss at the node + optimal cost of left subtree + optimal cost of right subtree, if they inherit j copies

cost[1, 1] = 0 + 0 + 1 = 1cost[1, 2] = 1 + 1 + 0 = 2cost[2, 1] = 1 + 0 + 1 = 2

cost[2, 2] = 0 + 1 + 0 = 1

cost[1] = 1cost[2] = 1

cost[1, 1] = 0 + 1 + 1 = 2cost[1, 2] = 1 + 0 + 1 = 2cost[1] = 2

Step 2: Enumerate all histories from the cost tables

• Maintain 3 variables for each node

• dups = optimal number of duplicated genes

• losses = optimal number of lost genes

• out = optimal number of genes to pass to its children

out = 1, losses = dups = 0

dups = 1losses = 0

out = 1 , losses = dups = 0

dups = 0losses = 0

dups = 1losses = 0

Step 3: Build a gene tree to represent the history

• From step 2: 1 duplication in humans & 1 duplication in frogs

• Build the gene tree with this information & the topology of the species tree

Hybrid Model