Bioinformatics: Alignment & Similarity

• Bioinformatics II focuses on Alignment & Similarity.

Learning Objectives

• Sequence analysis encompasses alignment and similarity assessments.
• Alignment methods are employed to align sequences, and their effectiveness is evaluated through alignment scoring.
• Homology searches, particularly using BLAST, are conducted to identify related sequences.

Why Align Sequences?

• Sequence alignment enables the measurement of similarity between sequences.
• It allows for observation of regions of conservation and variability.
• It helps determine residue-residue correspondences and predict the function of new sequences, which is invaluable for genome annotation.

Sequence Alignment and Common Ancestry

• Sequence similarity often suggests a common evolutionary origin.
• Aligning sequences presumes a shared ancestor, indicating homology.
• Protein sequences evolve, and alignments pinpoint homologous positions.

Pairwise Alignment

• Pairwise alignment compares two sequences, usually visualized using a dotplot matrix.
• Dotplots offer an overview of similarities between two sequences, with rows and columns corresponding to residues in each sequence.

Dotplots

• Dotplots are simple matrix representing the similarity between sequences
• Rows and columns represent a sequence each
• In a dotplot, if two sequences are identical, a diagonal line appears.
• Differences appear as interruptions or shifts in the diagonal.

Gaps in Alignments

• Gaps are insertions or deletions introduced to improve alignment matching.
• Gaps influence alignment scoring.

Global Alignments

• Global alignments compare sequences in their entirety and are useful for short, similar sequences.
• An example tool is NEEDLE, employing the Needleman-Wunsch algorithm.

Local Alignments

• Local alignments identify regions of highest similarity and extend the alignment outward.
• Conserved regions in distantly related sequences are found through this method usually.
• The WATER tool uses the Smith-Waterman algorithm.

Multiple Sequence Alignment (MSA)

• Multiple sequence alignment (MSA) compares three or more sequences.
• CLUSTAL is an example of MSA algorithms.

Features of Sequence Alignment

• Accession numbers provide an entry point into a database
• Gaps represent insertions or deletions within a sequence alignment.
• Consensus symbols, such as asterisks, colons, and full stops, denote levels of conservation

Amino Acid Properties

• Asterisks indicate fully conserved residues, colons indicate conservation between strongly similar groups, and periods indicate conservation between weakly similar groups

Alignment Scoring

• Alignment scoring is essential and influenced by:
• Sequence composition
• The aim is to score the similarities and differences between sequences
• Gaps are often penalized, as this can affect the alignment score

Percentage Sequence Identity

• Percentage sequence identity indicates variance.
• Gaps (insertions/deletions) affect sequence identity.

Complex Scoring Systems

• Protein sequence alignment algorithms use PAM or BLOSUM.
• Numerical scores are derived from substitution matrices to specific amino acid pairings.
• Known mutations impact the score derived from substitution matrices when scoring
• Biochemical similarity and the probabilty of occurance are also factors
• Score is based on what alignment is using for the process

Gaps and Penalties

• Gaps must be accounted for, with "gap penalties" used to avoid unchecked gaps and nonsense alighments.
• Gap creation is penalized with negative scores.

BLAST

• Basic Local Alignment Search Tool locates regions of local similarity between nucleotide or amino acid sequences.
• BLAST enables comparison of a query sequence against a database, identifying similar sequences by finding regions of similarity with local alignments.
• You can identify evolutionarily conserved regions
• You can find regions of common structure and function
• BLAST can find evolutionarily conserved regions for structure and/or function.
• These regions are often strictly required.

BLAST Results

• It identifies unknown sequences and find homologous sequences.

Types of BLAST Searches

• BLASTP: Amino acid sequence against protein sequences.
• BLASTX: Translated nucleotide sequence against protein sequences.
• TBLASTN: Amino acid sequence against translated nucleotide sequences.
• TBLASTX: Translated nucleotide sequence against translated nucleotide sequences.
• BLASTN: Nucleotide sequence against nucleotide sequences.
• PSI-BLAST is the most sensitive program and useful for finding very distantly related proteins

Key Summary Points

• Sequence similarity and alignment is central to bioinformatics analysis
• Global alignments are best for short and similar sequences
• Local alignments are used for finding regions of homology
• A BLST tool assists in finding unknown sequences, or evolutionarily conserves regions.
• Choosing a BLAST program depends on use case.