Biology · Genomics, Omics & Bioinformatics
Sequence alignment, homology and comparative genomics
ACGTAC aligned with ACG-AC has five matching columns and one gap. Is the identity one hundred percent or about eighty-three percent? Both figures can describe this same displayed alignment if their denominators differ. Neither percentage alone supplies the sequences' evolutionary history.
Gene Expression: RNA & Proteins explains how sequence, RNA and protein are related. Phylogeny, Classification & the Tree of Life introduces ancestry and the interpretation of trees. Research Design, Data & Scientific Practice explains sampling, independent units and the limits of comparisons. Their English chapters are available in this working manuscript; publication remains pending. The examples here use letter strings, inclusive coordinates, averages, fractions and ordinary graph reading. Each convention is introduced before it is used.
Back to the Biology mapBy the end, you can
Calculate an explicitly defined alignment comparison and distinguish measured similarity from inferred ancestry and functional prediction.
An alignment proposes corresponding positions
An alignment arranges sequence positions into proposed correspondences. A gap represents a position without a paired symbol in that arrangement; it is not an observed nucleotide. Matching letters, mismatches and gaps contribute differently to a chosen scoring rule. Different rules, boundaries or local versus whole-sequence comparisons can change the displayed result.
The original alignment has six columns. Five contain paired matching letters and the fourth contains T opposite a gap. Counting matches over all alignment columns gives five divided by six. Counting only columns with two letters gives five divided by five. A percentage becomes interpretable only when the aligned region and denominator are stated.
For its simple assigned score, a match contributes two, a paired mismatch minus one and a gap column minus two. The displayed arrangement scores eight. This arithmetic checks the chosen arrangement, not whether an algorithm would select it as optimal or whether its score is surprising in a biological database.
Identity changes with the denominator
Read the diagram in words
Columns one, two, three, five and six match. Five/six is all-column identity; five/five is paired-letter identity. Five times two minus two gives score eight.
| Alignment column | First assigned string | Second assigned string with a gap | Two letters in this column | Matching paired letters | Assigned column score |
|---|---|---|---|---|---|
| 1 | A | A | ✓ | ✓ | 2 |
| 2 | C | C | ✓ | ✓ | 2 |
| 3 | G | G | ✓ | ✓ | 2 |
| 4 | T | - | × | × | -2 |
| 5 | A | A | ✓ | ✓ | 2 |
| 6 | C | C | ✓ | ✓ | 2 |
| Matching paired letters ✓ | 2 |
|---|---|
| Matching paired letters × | -1 |
| Gap column | -2 |
| Matches divided by all columns | 5/6 = 0.8333333333333334 |
|---|---|
| Matches divided by paired-letter columns | 5/5 = 1 |
| Sum of assigned scores | 8 |
No ancestry or function is supplied by this score
Original tiny DNA-letter alignment supplied as a fixed arrangement. Gap is a display symbol, not a nucleotide. Assigned match/mismatch/gap-column weights are not a biological substitution model, optimality proof, search statistic or homology probability.
Homology concerns shared ancestry
Similarity is a property of a comparison. Homology is a relationship of common evolutionary ancestry inferred from evidence. A statement that two regions have eighty percent identity is meaningful under a defined alignment; calling them eighty percent homologous confuses the comparison with the ancestry claim.
Appropriately assessed excess similarity can support common ancestry. Its evidential meaning depends on sequence length, composition, the comparison and the search context. A short identical pattern can appear without establishing a shared gene history. Conversely, failure to detect a strong sequence match does not establish that common ancestry is impossible.
The displayed toy letters have no assigned real organism or history. Their identity calculation therefore remains a comparison only. For the separate genealogy model, ancestry is stipulated explicitly so that we can practise reading the relationships. It is not inferred from the six-column alignment.
Gene relationships are not a function guarantee
In the assigned genealogy, an ancestral copy duplicates before the lineage splits into species A and B. Each species retains two descendants. A1 and B1 are related through the first copy's speciation event; A1 and A2 are related through the earlier duplication. Orthology and paralogy describe these event relationships.
Genes in different species can also be paralogues: A1 and B2 trace their most recent shared gene ancestor to the duplication. Species membership alone therefore does not identify the relationship. The model preserves the two duplication-derived branches so that rotating the drawing would leave these relationships unchanged.
Related sequences can provide clues to structure or function, but a particular activity needs additional evidence and context. A conserved region may support a narrower comparison than an entire protein; related products can differ in regulation or specificity. Comparative genomics combines sequence relationships with other evidence rather than treating the closest-looking string as a complete functional answer.
The shared gene ancestor determines the relationship
Read the diagram in words
The duplication has branches to two copy-specific speciation nodes. A1/B1 and A2/B2 share speciation ancestors; A1/A2 and A1/B2 share the duplication ancestor.
| Assigned ancestral event or descendant | Gene relationship |
|---|---|
| Ancestral duplication | Copy divergence through duplication |
| First copy at species split | Copy divergence through speciation |
| Second copy at species split | Copy divergence through speciation |
| Assigned ancestral event or descendant | Species identity |
|---|---|
| Species A, copy one · A1 | A |
| Species B, copy one · B1 | B |
| Species A, copy two · A2 | A |
| Species B, copy two · B2 | B |
| Assigned ancestral event or descendant |
|---|
| Ancestral duplication → First copy at species split |
| Ancestral duplication → Second copy at species split |
| First copy at species split → Species A, copy one |
| First copy at species split → Species B, copy one |
| Second copy at species split → Species A, copy two |
| Second copy at species split → Species B, copy two |
| Assigned ancestral event or descendant | Gene relationship | Most recent shared gene ancestor |
|---|---|---|
| A1 ↔ B1 | Orthologues through speciation | First copy at species split |
| A2 ↔ B2 | Orthologues through speciation | Second copy at species split |
| A1 ↔ A2 | Paralogues within species A | Ancestral duplication |
| A1 ↔ B2 | Paralogues across species | Ancestral duplication |
History is assigned independently of the toy alignment
Original assigned gene history: duplication before species divergence, two retained branches in each species, no loss, transfer, recombination, dates or sequence-derived tree inference. Coordinates are layout only; no functional identity follows.
Worked example
Separate an arithmetic comparison from an ancestry account
The aligned strings are ACGTAC and ACG-AC. Matching columns are one, two, three, five and six. The all-column identity is five/six, approximately 0.8333, while paired-letter identity is five/five, or one. The gap must stay visible beside either summary.
Five matches contribute ten score units and one gap contributes minus two, giving eight. No random reference distribution or database-search statistic was supplied. The score cannot be converted into a probability of common ancestry or identical function.
In the separate assigned tree, A1/B1 and A2/B2 are orthologous pairs. A1/A2 and A1/B2 are paralogous pairs because their shared ancestor is the duplication. These statements follow from the stipulated events, not from branch spacing, species names alone or an identity threshold.
Check your understanding
Why do the two identity percentages disagree without an arithmetic error?
Show reasoning
One divides five matches by all six columns; the other divides by the five columns with two letters. The summaries use different denominators and must retain the gap and convention to be meaningful.
Does an identical two-letter pattern establish homology or function?
Show reasoning
No. Its identity can be calculated, but such a short comparison lacks sufficient context for that biological inference. Common ancestry and specific activity require suitable additional evidence.
Can A1 and B2 be paralogues even though they are in different species?
Show reasoning
Yes. In the assigned genealogy, their shared gene ancestor is the duplication before speciation. Different species alone do not make a gene pair orthologous.
What to carry forward
State the alignment and denominator, then separately justify any ancestry or function inference.
References
Resources supporting this lesson’s explanations and examples.