DNA mutation generator
v2.1.0Mutate DNA online by exact count or mutation rate, with reproducible controls and FASTA export.
Mutate DNA online by exact count or mutation rate, with reproducible controls and FASTA export.
The DNA mutation generator creates reproducible mutated versions of one or more DNA sequences. Paste a plain sequence or FASTA, choose substitutions, insertions, deletions, or a random mixture, then copy or download the generated variants as FASTA.
The calculation runs in your browser. The same input, settings, and random seed produce the same variant batch. Every seed from 0 to 999999 starts a separate random stream. Batches made before September 25, 2026 used an older random number generator, so their seeds no longer reproduce them.
>example_gene
ATGGCTTACGACTAAWith three variants, two substitutions per variant, all substitutions allowed, seed 42, and no line wrapping, the result is:
>example_gene|variant_001
ATGACTTACCACTAA
>example_gene|variant_002
ATGGATTACGAGTAA
>example_gene|variant_003
ATGGATTACAACTAAThese sequences are generated by the tool's deterministic mutation logic. Changing the seed creates a different batch while keeping the other settings fixed.
| Input | Accepted value |
|---|---|
| Plain DNA | A sequence containing only A, T, C, and G |
| FASTA text | One to five records with headers beginning with > |
| File upload | .txt, .fasta, .fa, .fas, or .seq |
| Maximum file size | 50 MB |
Whitespace and letter case are normalized. Ambiguity codes such as N, R, or Y are not accepted because the generator needs an unambiguous original base for every mutation.
| Setting | What it controls | Range or choices |
|---|---|---|
| Variants per input sequence | Number of output sequences created from each source | 1 to 100 |
| Mutation amount | How the number of events is determined | Exact event count or percentage of input length |
| Mutations per variant | Exact mutation events applied to each output | 1 to 50 |
| Mutation rate | Input length multiplied by a percentage, rounded to the nearest event | 0.1% to 50%, with 1 to 50 events |
| Variant preset | Allowed mutation types | Substitutions, insertions, deletions, or mixed |
| Substitution bias | Allowed replacement bases | All, transitions, transversions, A/T only, or G/C only |
| Insertion/deletion size | Bases added or removed by each indel event | 1 to 20 |
| Protect first 3 bases | Keeps the first three bases unchanged | On or off |
| Protect last 3 bases | Keeps the final three bases unchanged | On or off |
| Random seed | Makes a batch reproducible | 0 to 999999 |
| Require unique variants | Skips duplicate output sequences | On or off |
| FASTA line wrapping | Controls output line width | None, 60, or 80 characters |
| Output headers | Preserves the source name or uses indexed names | Two header formats |
Exact event count applies the selected number of mutation events to every variant.
Percentage of input length calculates the event count separately for each source sequence:
mutation events = round(input length × mutation rate / 100)The result is limited to at least one and at most 50 events. A 200-base sequence at 5% therefore receives 10 mutation events per variant. This setting describes simulated mutation events, not an experimentally measured mutation frequency in a cell population.
Enable Protect first 3 bases to avoid changing a possible start codon. Enable Protect last 3 bases to avoid changing a possible stop codon. The protected bases are excluded from substitutions and deletions, and insertions are placed outside the protected three-base blocks.
For coding sequences, these controls preserve the terminal triplets only. They do not verify the reading frame, identify the actual coding region, or guarantee that internal mutations preserve protein function.
Each selected position receives a different nucleotide. The bias setting controls the possible replacements:
| Bias | Allowed changes |
|---|---|
| All substitutions | Any base can change to any of the other three bases |
| Transitions only | A ↔ G and C ↔ T |
| Transversions only | Purine ↔ pyrimidine changes |
| A/T only | The replacement is A or T and differs from the original |
| G/C only | The replacement is G or C and differs from the original |
Positions are unique within a substitution-only variant, so a base is not selected twice in the same output.
Each event inserts a random DNA segment of the chosen indel size after a selected position. Insertion positions are unique within an insertion-only variant.
Each event removes a segment of the chosen indel size. Deletion-only variants use non-overlapping segments so the requested events have distinct source locations.
Each event randomly chooses among the currently possible substitution, insertion, and deletion operations. Since indels change sequence length, later events operate on the sequence produced by earlier events.
The results show:
Substitutions and inserted bases can be highlighted in the final sequence. Deleted bases no longer exist in the output sequence, so their locations are reported in the mutation records rather than highlighted as output bases.
| Goal | Tool |
|---|---|
| Introduce controlled random changes into an existing sequence | DNA mutation generator |
| Create a new random sequence with a chosen length and composition | Random DNA generator |
| Reorder the bases of an existing sequence while preserving composition | DNA shuffler |
| Reverse and complement a DNA sequence | Reverse complement |
| Remove digits, spaces, and non-DNA characters from a sequence | Filter DNA sequences |
| Convert a protein sequence into possible coding DNA | Protein to DNA converter |
A DNA mutation is a change in a nucleotide sequence. Mutations can arise during DNA replication, through imperfect repair, or after exposure to mutagens. They are a source of genetic variation, but their biological effect depends on the sequence context, affected gene, cell type, and organism.
This generator simulates nucleotide-level sequence changes. It does not predict whether a generated variant is viable, pathogenic, beneficial, expressed, or structurally stable.
A point mutation changes one nucleotide. A transition swaps bases within the same chemical class, while a transversion swaps a purine and a pyrimidine.
In a protein-coding region, a substitution may have several outcomes:
For example, changing the coding DNA codon TCA to TCG still encodes serine, while changing GAG to GTG changes glutamate to valine.
The generator does not select substitutions by their protein-level outcome. Translate and inspect generated coding variants before treating them as synonymous, missense, or nonsense designs.
An insertion adds one or more nucleotides:
Original: ATGCATGGCTAA
Inserted: ATGCAATGGCTAA
+In a coding region, an insertion whose length is not a multiple of three shifts the downstream reading frame. An insertion of three, six, or another multiple of three bases adds whole codons but may still disrupt protein structure or function.
A deletion removes one or more nucleotides:
Original: ATGGCATGCTAA
Deleted: ATGGATGCTAA
-As with insertions, a deletion that is not a multiple of three changes the downstream reading frame in a coding sequence.
Codons are read in groups of three. Adding or removing a number of bases that is not divisible by three shifts those groups:
Original codons: [ATG][GCA][TGC][AAG][TAA]
After deletion: [ATG][GCT][GCA][AGT][AA...]A frameshift can change every downstream amino acid and may introduce an early stop codon. To model in-frame indels, choose an indel size that is a multiple of three.
The same nucleotide change can have different consequences depending on:
Generated variants are hypotheses for downstream analysis or experimental design, not evidence of a biological phenotype.
When designing coding sequences, mutation type and placement matter. Synonymous substitutions can change codon usage or sequence composition without changing the encoded protein, while nonsynonymous changes may alter protein behavior. Indels can add or remove amino acids or change the entire reading frame.
Protecting the first and last three bases can keep terminal codons intact, but additional checks are still needed. Consider translating each output, screening motifs, checking GC content, and evaluating the protein sequence. The Protein to DNA converter is better suited to generating coding DNA from a desired protein sequence.
Not with this tool. It mutates DNA bases directly and does not constrain replacements to synonymous codons. Use the results as candidate sequences and translate them to confirm their protein-level effects.
No. It is a convenient way to scale the number of simulated mutation events with input length. It does not model replication cycles, selection, population size, or a biological error process.
Yes. Enable protection for the first and last three bases. Confirm that those triplets are the actual terminal codons of your coding region before relying on the setting.
When unique variants are required, a short sequence or restrictive settings may not provide enough distinct outcomes. The results include a warning instead of repeating variants.
The mutation calculation runs client-side in your browser. You can use the tool without creating a compute job.
Optimize coding DNA for a target organism with DNA Chisel while preserving its translation.
Generate random DNA sequences with controlled length, GC content, and restriction sites.
Generate random protein sequences with controlled length, amino acid composition, and motifs.
Generate random RNA test sequences with adjustable GC content and optional RNA-like templates and sequence motifs.
Shuffle DNA sequences while preserving nucleotide, dinucleotide, or k-mer composition.
Generate ligand protonation states over a pH range with Dimorphite-DL.
Design overlapping peptide libraries with adjustable length, overlap and terminal restrictions. Runs in your browser.
Enumerate stereoisomers from SMILES while preserving assigned stereochemistry by default.
Convert CSV or TSV sequence tables to FASTA with automatic delimiter detection and column mapping.
Translate DNA sequences into proteins across reading frames and genetic code tables.