bioinformatics lecture 4

What is dynamic programming in sequence alignment?

A method that checks all possible alignments to find the best one 🎯🧮

Why is dynamic programming considered accurate but slow?

It guarantees the optimal alignment but takes a lot of time ⏳⚠️

What type of alignment does Needleman–Wunsch perform?

Global alignment (aligns entire sequences end-to-end) 🌍➡️🌍

What type of alignment does Smith–Waterman perform?

Local alignment (finds best matching regions only) 🔍📍

What is the key difference between global and local alignment?

Global aligns everything, local finds the best matching parts 🧠✂️

Why does Smith–Waterman reset negative scores to zero?

To allow local regions of similarity to stand out 🔄⭐

How are alignments represented in dynamic programming?

As paths through a scoring matrix 🧩📊

Why can multiple optimal alignments exist?

Different paths can give the same best score 🛣️🏆

What three components are needed to score alignments?

Match/mismatch scores, gap penalties, and substitution matrices 🧮📐

Why are dynamic programming algorithms unsuitable for databases?

They are too slow for millions of sequences 🐢💾

What is a heuristic alignment algorithm?

A faster method (than dynamic programming) that sacrifices perfect accuracy for speed ⚡🎯

Why are heuristic algorithms needed?

To search large databases quickly 🚀🗂️

What is BLAST? (Basic Local Alignment Research Tools)

A fast tool for finding local sequence similarity 🔍🧬

What kind of alignment does BLAST perform?

Local alignment only 📍🧬

What is the core idea behind BLAST’s speed?

It searches only promising regions, not everything ⚡🧠

What is the steps in BLAST?

1. Break query into short “words”

2. Find exact word matches in database

3. Extend matches into longer alignments

4. joins shorter alignments which Allow mismatches and gaps

Why are low-complexity regions filtered before BLAST?

Repeats can cause false matches ⚠️🔁

What is the raw score in BLAST?

A score based on matches minus penalties ➕➖

What does a higher raw score mean?

A better alignment 👍

What is a bit score in BLAST?

A normalised score that allows comparison across searches ⚖️📊

Why is the bit score useful?

It is a normalised measure that allows you to compare the alignment and It is independent of database size 📦❌

What is query coverage in BLAST?

The percentage of the query sequence that aligns 📏📊

What is the E-value in BLAST?

The number of matches expected by chance 🎲📉

How should E-values be interpreted?

Smaller E-value = more significant match ⭐⬇️

What is the rule of thumb for E-values?

E ≪ 1 is meaningful; E ≈ 1 is likely random 🚦

Why must BLAST results be interpreted using multiple values?

Score, coverage, and E-value give different information 🧠📊

What does BQE stand for in BLAST interpretation?

Bit score (strength), Query coverage (how much), E-value (significance) 🅱️🆀🅴

What is multiple sequence alignment (MSA)?

Aligning more than two sequences at once 👥🧬

Why are MSAs more informative than pairwise alignments?

They reveal conserved and important regions 🔎⭐

What is a conserved region in an alignment?

A region that stays similar across sequences 🔒🧬

Why are conserved regions important?

They are usually functionally important 🧠⚙️

What do variable regions indicate?

Flexible or non-essential regions 🔄🧬

Why must MSAs include both close and distant sequences?

Close gives signal and shows what important, distant gives variation and shows what’s not important 🎯🌍

What is the main strategy used by Clustal Omega?

Progressive alignment ( by progressively adding alignments in order of similarity) 🌳➡🧬

What are the main steps of progressive alignment?

Pairwise comparison, guide tree, step-by-step alignment 🪜🌳

What is a key limitation of progressive alignment?

Early mistakes cannot be fixed ❌🔒

Why are MSAs not guaranteed to be optimal?

They use heuristics, not exhaustive searching ⚠️🧮

What is a consensus sequence?

The most common residue at each position 📊🧬

Why is a consensus sequence NOT an ancestral sequence?

It reflects frequency, not true evolutionary history ❌🌳

What can bias a consensus sequence?

Over-representation of certain species 🧠⚖️

Why can consensus sequences be misleading?

They may reflect sampling bias, not biology ⚠️📉