Sanger Sequence Assembly Online - Join Forward and Reverse Reads
Assemble .ab1 or .scf reads into one contig with no reference sequence - the forward/reverse pair of an insert, or a set of tiling reads.
🔒 Nothing you paste is logged or stored on our servers
Every other Sanger tool here compares a read against something you already have. This one is for the other half of the job: you sequenced something new and do not have its sequence yet. Upload the forward and reverse traces of one insert and they are oriented, trimmed and merged into a single consensus, with per-position depth and the positions where the reads disagree listed separately - because a difference two reads agree on and a difference one read reports are not the same finding. Reads that overlap nothing are reported as unassembled rather than glued on. What you will not find in the output is a combined Phred score for the consensus: adding two Phred 40 bases does not make a Phred 80 base, so what comes back is depth, agreement, and the best single observed quality at each position - three real measurements instead of one invented one.
Parsed in your browser — only the base calls and quality scores are sent. Pick the forward and reverse reads together.
0 pasted reads. A pasted read has no quality scores, so its ends are not trimmed. Uploaded traces take precedence.
Upload two or more traces, or paste the reads, to assemble them into one contig.
How to use the Sanger Assembly tool
- 1Upload your trace files (.ab1, .abi or .scf), or paste the reads as a multi-record FASTA. Traces are parsed in your browser and their quality scores are used to trim the ends.
- 2Leave the overlap settings alone unless the assembly fails - 30 bases at 85% identity is a sensible floor for Sanger.
- 3Check whether everything joined into one contig. That is the question most people are actually asking.
- 4Read the disagreements before you trust the consensus. Each one is either a real variant or a basecalling error, and only your reads can say which.
Frequently asked questions
Do I need a reference sequence?
No, and that is the point. This assembles the reads against each other. If you do have a reference, Sanger vs Reference and the clone screening tool will give you a verdict against it instead, which is a different and usually more useful question.
How does it know which read is reverse?
From the overlaps. With no reference there is nothing to vote against, so each read joins the growing contig in whichever orientation actually overlaps it. The result says which reads were reverse-complemented.
What happens to a one-base insertion in a single read?
It is reported, not swallowed. Reads that span the position vote against it, so a base only one read saw is outvoted and left out of the consensus while still appearing in the disagreement list. That case is worth calling out because a homopolymer miscall is the commonest Sanger error there is.
Why is there no quality score on the consensus?
Because there is no honest way to compute one. Summing Phred scores does not produce a Phred score, and a number shaped like one would travel into every downstream tool that reads quality. Depth, agreement and the best observed Phred per position are reported instead.
Is my trace file uploaded anywhere?
The trace is parsed in your browser and only the base calls and quality arrays are sent for assembly. Nothing is logged or stored.
More
Related tools
Decompose one messy chromatogram into fractions over the molecules your reaction could have produced - not onto a generic indel ladder.
Re-read a chromatogram for the second peak under each call, and for the bases past where the basecaller gave up.
Align several DNA or protein sequences and view a colored alignment with a consensus.