Written and maintained by CASRAI Editorial Board
Last updated
An open reading frame (ORF) is, in the words of the NHGRI Talking Glossary, “a portion of a DNA sequence that does not include a stop codon.” In practice, researchers use ORF searches to find stretches of a new sequence that could encode a protein. The free NCBI ORFfinder web tool does this for you and returns the range of each ORF along with its protein translation. This guide explains what an ORF is used for and walks through ORFfinder’s actual settings, showing how each one changes the result and how to read the output table.
What it is used for in research
The NCBI ORFfinder page describes the tool as one that “searches for open reading frames (ORFs) in the DNA sequence you enter.” A related search tool, the Sequence Manipulation Suite ORF Finder, states its use case directly: to “search newly sequenced DNA for potential protein encoding segments.” Typical uses therefore sit early in a sequence-analysis workflow: you have a new nucleotide sequence, you want candidate protein-coding regions, and you then take the translated proteins into a similarity search. For the wider field this work belongs to, see what bioinformatics is and what genomics covers.
Install or access (web tool, no install)
ORFfinder runs in the browser at ncbi.nlm.nih.gov/orffinder. Nothing is installed. The page asks you to “enter accession number, gi, or nucleotide sequence in FASTA format.” The web version accepts a sequence subrange of up to 50 kb, so a larger sequence has to be analysed in pieces by choosing subrange coordinates.
Step-by-step in ORFfinder
- Enter the sequence. Paste a FASTA sequence or an accession number into the input box on the ORFfinder page.
- Set the subrange (optional). Enter start and end coordinates if you only want part of the sequence searched. The web version is limited to a subrange of up to 50 kb.
- Choose the minimal ORF length. The menu offers 30, 75, 150, 300 or 600 nucleotides. The tool reports ORFs “equal or more than the selected value.”
- Choose the genetic code. The numbered list starts with “1. Standard” and also includes mitochondrial and plastid codes, among others. Pick the one that matches the organism and compartment your sequence comes from.
- Choose the ORF start codon. The options are “ATG” only (the default), “ATG” and alternative initiation codons (as defined by the selected genetic code), or any sense codon.
- Decide on nested ORFs. Tick the option to ignore nested ORFs if you want to skip ORFs that lie completely inside another ORF.
- Submit and read the results. The tool returns the range of each ORF with its protein translation, as described in the next section.
What each setting changes
| Setting | Options on the NCBI page | Effect on the result |
|---|---|---|
| Minimal ORF length | 30, 75, 150, 300, 600 nt | Only ORFs equal to or longer than the selected value are reported. |
| Genetic code | Numbered list from “1. Standard”, including mitochondrial and plastid codes | Sets the codon table used for the search and the protein translation. |
| Start codon | ATG only (default); ATG and alternative initiation codons; any sense codon | Controls what counts as the beginning of an ORF. The “any sense codon” option finds all stop-to-stop ORFs. |
| Ignore nested ORFs | On or off | Leaves out ORFs that are completely inside another ORF. |
Output and how to read it
The results table lists, for each ORF, a Label, Strand, Frame, Start, Stop and Length (given in nucleotides and amino acids). Marked ORFs are highlighted in the table and framed on the graphic in the Sequence Viewer. You can work with the results in several ways:
- Six-frame translation: “Display six-frame translation” opens a draggable widget, with the options “Only ORFs translation” and “Six full length translations.”
- Mark a subset: the Mark subset menu lets you select Strand(+) or Strand(-) ORFs.
- Download: “Download marked set” exports Protein FASTA, CDS FASTA, ASN.1 or a Feature table.
- Search for similar proteins: the active ORF’s protein can be sent to SmartBLAST or BLAST, and a marked set can go to protein BLAST against UniProtKB/Swiss-Prot, Reference proteins (refseq_protein) or Non-redundant protein sequences (nr).
The Strand and Frame columns are the quickest way to see where each ORF sits. Length matters when you compare candidates: a larger minimal length setting removes short hits before you ever see them. For follow-up on similarity searching, see how to use NCBI Primer-BLAST, a separate NCBI tool for designing primers once you know the region you want to amplify.
Troubleshooting
- No ORFs, or fewer than expected: lower the minimal ORF length, since ORFs shorter than the chosen value are not reported.
- Expected ORF is missing: check the start codon option. With “ATG” only, ORFs beginning with an alternative initiation codon are not reported; switch to the alternative-initiation option or to “any sense codon” to see stop-to-stop ORFs.
- Translation looks wrong: confirm the genetic code. Mitochondrial and plastid sequences have their own entries in the genetic code list.
- Sequence too long: the web version is limited to a subrange of up to 50 kb, so split the sequence and run each part.
- Too many overlapping ORFs: turn on the option to ignore nested ORFs.
Free alternatives
| Tool | Type | What its own page says |
|---|---|---|
| Sequence Manipulation Suite ORF Finder | Web page | Searches the DNA sequence you enter for ORFs, returns each ORF’s range with its protein translation, and supports the entire IUPAC alphabet and several genetic codes. |
| orfipy | Python/Cython tool | Described as a tool to extract ORFs in a fast and flexible manner, particularly fast for data with many smaller FASTA sequences such as de novo transcriptome assemblies or collections of microbial genomes. |
| ORFik | R package (Bioconductor) | An R package for analysis of transcript and translation features using sequence data and NGS data such as Ribo-Seq and RNA-Seq; its page lists open reading frames among its uses. Licence listed as MIT + file LICENSE. |
Related molecular-biology workflows on CASRAI include site-directed mutagenesis, where knowing the coding region of a sequence matters when you design changes to it.
FAQ
What is an open reading frame?
The NHGRI Talking Glossary defines it, as related to genomics, as a portion of a DNA sequence that does not include a stop codon, the codon that functions as a stop signal.
How do I find open reading frames in a sequence?
Paste the sequence or accession number into NCBI ORFfinder, set the minimal length, genetic code and start codon, and run it. The tool returns each ORF’s range with its protein translation.
What does “any sense codon” do in ORFfinder?
It lets an ORF start at any sense codon, which finds all stop-to-stop ORFs rather than only those that begin with a start codon.
Can I export the proteins from ORFfinder?
Yes. “Download marked set” offers Protein FASTA, CDS FASTA, ASN.1 and Feature table formats.
Sources (official docs)
- NCBI, ORFfinder tool page, ncbi.nlm.nih.gov/orffinder (checked 2026-10-29).
- NCBI, ORFfinder documentation, ncbi.nlm.nih.gov/orffinder/docs (checked 2026-10-29).
- NHGRI, Talking Glossary: Open Reading Frame, genome.gov/genetics-glossary/Open-Reading-Frame (checked 2026-10-29).
- Sequence Manipulation Suite, ORF Finder, bioinformatics.org/sms2/orf_find.html; orfipy on GitHub; ORFik on Bioconductor (all checked 2026-10-29).
