Lecture 11a - Phylogenetics in Bioinformatics

Lecture 11a - Phylogenetics - Bioinformatics

• Phylogenetics is the study of evolutionary relationships, often represented by "trees" or cladistics.
• Phylogenetic analysis infers evolutionary relationships.
• Clades are groups of organisms sharing a common ancestor.
• Members of a clade are closely related.

Learning Objectives

• Understand phylogenetic tree elements.
• Know assumptions made when interpreting phylogenetic trees.
• Differentiate between rooted and unrooted trees.
• Define and understand the relationships between homolog, ortholog, paralog, and xenology.
• Describe phylogenetic tree construction steps.

Phylogenetics

• Evolutionary theory forms the basis for assumptions in BLAST, multiple protein alignments, and sequence distance measurements.
• Identifying gene families, functions, origins of diseases, and polymorphisms are important applications.
• Best BLAST matches aren't always the most similar sequences.
• Misinterpretations of phylogenetic data are possible.

The Basics of Phylogenetics

• Phylogenetics studies evolutionary relationships.
• Phylogenetic analysis estimates these relationships, often using phylogenetic trees.
• Cladistics is sometimes used to indicate phylogenetic relationships.
• A clade is a group of organisms with a common ancestor.
• Members of a clade are more closely related to each other than to members of other groups.

Assumptions of Cladistics

• Organisms are related through descent from a common ancestor.
• Evolution follows a branching pattern (cladogenesis).
• Characteristics change over time within lineages.

Phylogenetic Trees - Elements

• Clades include the most recent common ancestor (MRCA) and all its descendants.
• Taxons are named groups of organisms.
• Branch lengths may not directly reflect evolutionary distances but always signify divergence.
• Nodes are branching points.

Phylogenetic Tree Example (and Data)

• Specific examples and data are provided in the examples. Please refer to the images for this data.

Assumptions of Molecular Phylogenetic Methods

• Sequences must be correct.
• Sequences are homologous, descending from a common ancestor.
• Each position in a multiple sequence alignment (MSA) is homologous to others.
• Multiple sequences in an alignment share a common phylogenetic history.
• Sampling of taxa is adequate.
• Sequence variation reflects the broader group.
• Sufficient sequence variability exists to resolve the issue.

More Assumptions

• Sequences may not always evolve according to one stochastic process.
• All positions in a sequence may not evolve according to the same process.
• Sequence positions often evolve independently.

Some Definitions

• Homologs share a common ancestor but may not have similar activities.
• Homology refers to shared ancestry, while similarity is quantifiable.
• Orthologs are homologs arising from speciation, usually exhibiting similar functions.
• Paralogs arise from gene duplication and often have differing functions.
• Xenologs are obtained from horizontal gene transfer, typically with similar function.

Rooted Phylogenetic Trees

• Rooted trees have a root node representing the common ancestor of all sequences/species.
• The root indicates evolutionary direction and ancestral/derived nodes/branches.
• Rooted trees help infer evolutionary time.
• Outgroups (more distantly related sequences/species) define the root in rooted trees.

Rooted Tree of Protein Domains (and Data)

• A specific example and data are presented in the examples. Please refer to the image for this data.

Unrooted Phylogenetic Trees

• Unrooted trees lack a root, thus not defining evolutionary direction or a common ancestor.
• They emphasize pairwise relationships and similarities/distances between sequences.
• Unrooted trees are common when the evolutionary path is unclear.

Unrooted Tree of a Gene Family (and Data)

• A specific example and data are presented in the examples. Please refer to the image for this data.

Rooted vs Unrooted Trees - Key Differences

• Rooted trees indicate common ancestry, unrooted trees do not.
• Rooted trees define an evolutionary direction, unrooted trees do not.
• Rooted trees depict ancestry/timelines, unrooted trees only show relationships based on similarity/distance.

How To Construct a Tree (Steps)

• Construct a multiple sequence alignment.
• Determine the substitution model.
• Build the tree.
• Evaluate the tree.

Alignment

• Start with a multiple sequence alignment of genes.
• Assume aligned positions (sites) share a common ancestor.
• Determine which base/gap represents the character state.
• Homologous sites are ones where evolution has occurred.
• Choose suitable software for alignment (e.g., ClustalW, ClustalX, T-Coffee, MALIGN).
• Subsets of the alignment can be used to construct the tree.

Alignment (Additional Details)

• Alignment editing (deletion/insertion of gaps) can be necessary.
• The goal is to create an alignment that best represents evolutionary processes.

Phylogenetic Data Model

• Substitution models describe how base rates change.
• Parsimony aims to find the simplest evolutionary tree.
• Weighted matrices (e.g., transition=1, transversion = 2) score distances between characters.
• Substitution matrices (e.g., PAM, BLOSSUM) reflect rate changes between characters derived from multiple sequence alignments.

Tree Building Methods

• Distance-based methods use sequence dissimilarities.
• Character-based methods use character data throughout the analysis.
• Include methods like UPGMA (unweighted pair group method with arithmetic mean) and NJ (neighbor joining).
• Other methods like Maximum Parsimony and Maximum Likelihood are used.

Tree Evaluation

• Assess tree correctness with Randomized Trees (Skewness Test) to analyze MP tree lengths distribution.
• Bootstrapping creates many datasets with substitution, to determine how often a specific branch is recreated.

Phylogenetic Software

• List of software like PHYLIP, PAUP, FASTDNAml, PUZZLE, and other tools such as Clustal, Treeview are part of phylogenetic analysis. Note: Specific names of software will vary.

This quiz covers the essentials of phylogenetics, focusing on evolutionary relationships and their representation through trees. Students will explore concepts such as clades, tree construction, and the various types of evolutionary relationships. It aims to enhance understanding of phylogenetic analysis in bioinformatics applications.