An ORFfinder Alternative Without the 50 kb Web Limit
NCBI ORFfinder is free and correct, so this page is not going to pretend it is a poor tool. It has one stated hard edge: the web version takes a subrange of up to 50 kb, and the documented way past that is a stand-alone binary for Linux x64. The ORF finder here has no fixed ceiling — a 250 kb contig scans in about 35 ms, in your browser, with no submission step.
Four rows of the table go the other way, and they are worth reading before you switch: the standard genetic code is the only one here, ORFs start at ATG and nowhere else, there is no BLAST step, and NCBI's record database is not behind the box. If your sequence is mitochondrial, or you need to identify what you found, ORFfinder is the right tool and this one is not.
SeqBench vs NCBI ORFfinder
| Aspect | SeqBench | NCBI ORFfinder |
|---|---|---|
| Access model | Web app — paste and the ORFs are already on screen, no submission step | NCBI web tool — submit the sequence and wait for the viewer |
| Price | Free | Free |
| Longest sequence it will take | No fixed ceiling — a 250 kb contig is scanned in roughly 35 ms, and the work is linear in the sequence, so a phage genome or a BAC goes straight in | The web version is limited to a subrange of up to 50 kb; past that the documented route is the stand-alone binary, which is Linux x64 only |
| Frames and strands | All six — three forward, three reverse, each ORF reporting its frame and strand | All six |
| Minimum ORF length | Any value you type, in amino acids, with no fixed steps | One of five fixed choices: 30, 75, 150, 300 or 600 nucleotides |
| Genetic code | No — the standard code only. If you are working on mitochondria, a ciliate or a plastid, ORFfinder is the correct tool and this one will give you the wrong protein | Yes — the standard code plus more than 25 others, including the mitochondrial and plastid codes |
| Alternative start codons | No — ORFs start at ATG. There is no alternative-initiation mode and no stop-to-stop mode, so a GTG- or TTG-initiated bacterial ORF is reported only from its first internal ATG | Yes — ATG only, ATG plus alternative initiation codons, or any sense codon for stop-to-stop ORFs |
| Nested and overlapping ORFs | Yes — every ATG in a frame gets its own entry rather than only the longest one in each span | Yes |
| Protein translation returned | Yes — each ORF with its coordinates, frame, strand, nucleotide and amino-acid length, and its translation | Yes |
| Checking a predicted protein against known ones | No — there is no similarity search here, so a predicted ORF is not identified, only found | Yes — SMART BLAST or BLASTP on any ORF, one click away |
| Starting from an accession | Not on this page — it takes pasted sequence or FASTA. Sequence can be fetched by accession with a separate tool and then pasted in | Yes — an accession, a GI, or pasted sequence, with the record database right behind it |
| Reading a file | Yes — FASTA and raw sequence here, and elsewhere on the site GenBank, SnapGene .dna and Sanger .ab1 files are parsed and the sequence carried over | FASTA or an accession |
| Where it runs | In your browser — the ORF scan is executed locally and the sequence is not uploaded anywhere | On NCBI's servers; the sequence is submitted |
| API and AI-agent access | Yes — a JSON REST endpoint and an MCP server, so an agent can call it directly | No web API for ORFfinder itself; the stand-alone Linux binary is the scriptable route |
Comparison is simplified and provided in good faith — check ncbi.nlm.nih.gov/orffinder for current behavior. ORFfinder, BLAST and SMART BLAST are services of the National Center for Biotechnology Information and are not affiliated with SeqBench.
Where SeqBench is the better fit
- Your sequence is over 50 kb and you do not want to install a Linux binary
- You want the answer as you paste, with no submit-and-wait step
- You want a minimum length that is not one of five fixed choices
- The sequence should not leave your browser
- You want ORF finding from a script or an AI agent without a key
Where ORFfinder still wins
- Any genetic code that is not the standard one — mitochondrial, plastid, ciliate
- Alternative initiation codons, and stop-to-stop ORFs
- Identifying what you found, with BLAST one click away
- Starting from an accession instead of a pasted sequence
- Sequences past 50 kb, via the stand-alone binary, if you run Linux
Related sequence tools
Frequently asked questions
My sequence is longer than 50 kb. What are my options?
That is the one case this page exists for. NCBI's own ORFfinder page states that the web version is limited to a subrange of the query sequence up to 50 kb long, and points at a stand-alone version for Linux x64 as the way past it — which is a fine answer if you have a Linux machine and a package manager, and no answer at all if you do not. The ORF finder here has no fixed ceiling: the scan is linear in sequence length and a 250 kb contig comes back in roughly 35 ms, in the browser, with nothing submitted anywhere.
Does it support the mitochondrial genetic code?
No. This is the standard code only, and it is the most important thing on this page to be clear about: running a mitochondrial, ciliate or plastid sequence through it will give you a protein that is wrong at specific codons rather than an error. ORFfinder offers more than 25 genetic codes and is the right tool for any of them.
Will it find ORFs that start at GTG or TTG?
Not as starts. ORFs here begin at ATG, so a bacterial gene initiated at GTG or TTG is reported from its first internal ATG instead — shorter than the real coding sequence, and with the N-terminal residues missing. ORFfinder has three modes for this: ATG only, ATG plus alternative initiation codons, and any sense codon for stop-to-stop ORFs. If alternative initiation matters to your organism, use it.
Does it find ORFs on the reverse strand?
Yes — all six reading frames, three on each strand, and every ORF says which frame and which strand it came from. Nested ORFs are kept too: every ATG in a frame gets its own entry rather than only the longest span, which matters when you are looking for an internal start site rather than the obvious gene.
Can I BLAST the proteins it finds?
Not from here. This tool finds ORFs; it does not identify them. ORFfinder puts SMART BLAST and BLASTP one click from every ORF, and for the question "what is this thing I just found?" that is a genuinely better workflow. What you can do here is carry the translation into the protein tools — properties, domain annotation, hydropathy — which answers a different question about the same sequence.
Is my sequence uploaded?
No. The ORF scan is one of the tools that runs entirely in your browser: the same code, executed locally, with no request leaving the page. That is also why there is no queue and no submit button — the answer appears as you paste.