See what readhit can do

readhit puts sequencing reads on one target gene and shows what they say about it.

The screenshots below show the main things it does:

Align Sanger and NGS reads to one reference
Place the reads with minimap2, in the browser
See coverage, depth and the shares of a column
Draw capillary reads as chromatogram traces
Open a SAM as the alignment it already is
Build a consensus and read it as protein

Files are processed in the browser, on your own machine — nothing is sent over the network.

Open readhit

A run on one gene, read as a pileupFour hundred amplicon reads on a 451 bp target: reads that do not overlap share a line, so the set takes a screenful instead of four hundred rows. Under the reference sit the depth profile and the consensus — the depth is the height of the bar, and the colours inside it are the bases that disagree. The reads are placed by minimap2 compiled to WebAssembly, in a worker, or by the built-in aligner.
Capillary reads with their tracesThree Sanger reads of one construct, on the construct’s own GenBank map. Each read is drawn as the trace it was called from, in the alignment’s coordinates, so a difference can be settled by looking at the peaks rather than at the letter somebody else called.
A SAM, opened as the alignment it isA hundred and fifty PacBio HiFi reads as minimap2 placed them, read straight out of the SAM: the file says where every read goes and readhit reads the CIGARs. Nothing is aligned or trimmed here — showing a picture the file does not contain would be the one thing a viewer must not do. The reads bring their own quality with them, and at Q93 a HiFi base is about as settled as a base gets.
Both methods, one targetOne specimen read twice — a capillary read and a deep amplicon run — on the same gene, in one session and one coordinate system. The column here is 252 reads deep and every one of them agrees; the row the readout names is the capillary read. Two methods answering the same question at the same position is the argument for keeping them together.
The consensus, and what it codes forThe reference and the consensus are translated in a chosen frame, under a chosen genetic code, each as its own lane. A difference is then read where it matters: the status bar names the codon it falls in and whether it changes the residue.