Sequence Identity Calculators

0 calculators tagged with “Sequence Identity”

Sequence identity is the percentage of positions in a sequence alignment where two sequences have the same nucleotide or amino acid. It is the most commonly reported metric for comparing biological sequences — used to assess similarity, infer evolutionary relationship (homology), predict protein function, and design primers. A sequence alignment must be performed before calculating identity, and the result depends on alignment parameters, gap penalties, and whether the comparison is global or local. Different thresholds of identity carry different biological implications.

All Calculators

No calculators found for this topic.

Calculating Percent Identity

% Identity = (Number of identical positions / Total aligned positions) × 100

Gaps can be excluded from the denominator (ungapped identity) or included (gapped identity). The choice affects the result — gapped identity is more conservative.

DNA Sequence Identity

Example alignment:
Sequence 1: ATCGATCGGTA
Sequence 2: ATCGTTCGGTA
Differences at position 5 (A vs. T): 10/11 identical → 90.9% identity

Protein Sequence Identity

Protein identity below DNA identity because synonymous codon changes don't alter amino acids. Interpretation thresholds (rough guidelines):

  • >40% identity: Likely structurally and functionally similar (confident homology)
  • 25–40% identity: Twilight zone — possible remote homology; need structural comparison
  • <25% identity: Very distant or no relationship; random sequence similarity threshold

BLAST (Basic Local Alignment Search Tool)

BLAST finds local sequence alignments between a query and database sequences:

  • BLASTN: Nucleotide query vs. nucleotide database
  • BLASTP: Protein query vs. protein database
  • BLASTX: Nucleotide query (translated) vs. protein database
  • TBLASTN: Protein query vs. translated nucleotide database

Key BLAST statistics: E-value (expected number of hits by chance — lower = more significant; <10⁻⁵ = significant); bit score; and percent identity.

Identity vs. Homology

These terms are frequently confused. Homology is a qualitative statement — two sequences either share common ancestry or they don't. Identity is a quantitative measure — a percentage. A pair of sequences is either homologous or not; saying '50% homologous' is incorrect usage. High identity is evidence of homology, but not proof.

Glossary

Sequence Identity
The percentage of identical positions in a sequence alignment: (identical positions / aligned positions) × 100. Used to assess similarity between DNA or protein sequences. Requires alignment before calculation.
BLAST (Basic Local Alignment Search Tool)
A widely used algorithm finding local sequence alignments between a query and database sequences. Key statistics: E-value (expected hits by chance; <10⁻⁵ = significant), bit score, and percent identity. Types: BLASTN, BLASTP, BLASTX, TBLASTN.
E-Value
The expected number of alignments with a given score or better that would occur by chance in a database search. Lower E-value = more statistically significant match. E-value < 0.001 is commonly used as a significance threshold in BLAST searches.

Frequently Asked Questions

% Identity = (identical positions / total aligned positions) × 100. Requires a sequence alignment first. Gaps can be excluded (ungapped identity) or counted against the score (gapped identity). Example: two sequences share 45 identical amino acids in a 60-position alignment (5 gaps in one sequence): ungapped identity = 45/55 = 81.8%; gapped identity = 45/60 = 75%. Always specify which method was used when reporting identity.

Generally: >40% identity suggests similar structure and function (confident functional inference); 25–40% is the twilight zone — possible remote homology, use structural analysis; <25% is at or below random similarity thresholds — relationship is uncertain. These are rough guidelines — enzyme active sites and critical structural elements can be conserved even at low overall identity. Context (known protein family, conserved motifs, structural similarity) improves functional inference.

BLAST finds local alignments between a query sequence and sequences in a database. It first finds short matching words (seeds), then extends them using a scoring matrix (BLOSUM62 for proteins, nucleotide matrices for DNA), allowing gaps. The E-value (expect value) = the expected number of hits with that score or better by chance in a database of that size. E-value < 0.001 is generally considered significant; < 10⁻⁵ is strong evidence of true similarity. Lower E-value = more significant match.

Sequence identity is a quantitative measure — the percentage of positions that match in an alignment. Sequence homology is a qualitative evolutionary statement — two sequences are homologous if they share common ancestry (descended from the same ancestral sequence). Homology is binary (homologous or not); identity is a percentage. High sequence identity is evidence of homology, but not proof. Saying two proteins are '70% homologous' is incorrect — they are either homologous or they are not.