Sequence alignment and reading a search result
Alignments receive scores for matches, mismatches and gaps under stated rules. BLAST can report an E-value: the expected number of chance alignments with that score or higher in the search space. Statistical support alone does not establish biological function.
What a learner can do afterwards
- Scores a short alignment by hand under a stated scheme, gaps included.
- Reads a search result and separates a strong hit from a chance one using the reported statistic.
- Explains why a short query can have weak statistical support even for a perfect match.
1 · Read
Assembly uses shared sequence; alignment makes a comparison explicit. Align DNA, RNA or protein sequences in columns and score matches, mismatches and gaps. Among candidates using the same scoring scheme and constraints, prefer the highest total. Reads can be aligned before a genome is assembled.
For our DNA examples, use +2 per matching column, -1 per mismatch and -2 per gap character. Count every displayed column, including end gaps. No column has two gaps. Other schemes can charge separately for opening and extending a gap, so read the stated rules.
Four matches, one mismatch and one gap character score 4 times 2, minus 1, minus 2: a total of 5. Count each category, multiply by its score and add the signed contributions. Prefer the higher total when candidates use the same rules and constraints.
BLAST reports an E-value: the expected number of chance alignments with that score or higher in the search space. It is an expected count, not the probability that a hit is false. Smaller values give stronger evidence against chance under the model. Check how much of the query aligns and whether the matched region supports a proposed function.
A short perfect match can still have a large E-value because few letters can match by chance. E = 5 means five alignments at least this strong are expected by chance in that search space. Perfect identity alone is then weak evidence. Length, composition, database size and scoring affect the statistic.
Use explicit scoring rules. An E-value reports expected chance matches; aligned length and biological evidence still matter.
Lesson, questions and flashcards written for this dataset.
2 · Watch
Take it off screen
Where it sits
8 questions wait behind this lesson, each with its answer explained. Every answer feeds the sky: stars light as they are learned, and dim when it is time to come back.