Guide track

Every guide the design returned, drawn where it binds along the construct, with its predicted cut site on a lane beneath — so clustering, and the gaps no PAM reaches, are visible rather than inferred from a table of coordinates.

What it doesCopy Anchor Link

A guide design returns a ranked list, and a ranked list answers "which guide is best" while hiding the question that usually decides an experiment: where can I cut at all? Guides are not evenly distributed. They exist where a PAM exists, so a construct has stretches dense with options and stretches with none, and a knock-in that has to land at a particular base may have no guide within reach of it however good the top of the list looks.

This is the positional view of the same result. The upper lane is every guide the panel is currently listing, drawn over the bases it binds and pointing the way it reads. The lower lane marks the predicted double-strand break for each one, which is a different coordinate from the guide's own span: for a Cas9 nuclease the break falls three bases inside the protospacer from the PAM, so a cut can land outside the feature the guide sits in. Seeing the two lanes together is what makes that concrete.

The highlight behind them is the region the design was scoped to. The track follows the panel's filters, so narrowing by score, by GC, or by off-target count changes what is drawn — which is the point: a filter that leaves a stretch of the construct empty has told you something about the construct.

When you'd use thisCopy Anchor Link

• To find the cut-free gaps. A homology-directed repair edit needs a break near the change, and the distance from the nearest available cut is a design constraint you can only see here.
• To check that a filter has not emptied a region. A stringent off-target threshold can remove every guide from the one place you need to cut.
• To choose a pair for a dual-guide excision. Two cuts define a deletion, and the deletion is the span between the two lower-lane marks, not between the two guides.
• To sanity-check the target scope. A guide drawn outside the highlight is a guide the design found beyond the region you asked about.

What this computesCopy Anchor Link

• The binding span and orientation of every listed guide, in the construct's own coordinates.
• The predicted cut site of each guide, from the nuclease in force, drawn as a point on its own lane rather than folded into the guide's span.
• The scoped region the design ran over, as a highlight behind both lanes.
• The selected guide, marked distinctly, so a row picked in the table is findable on the axis.

What it does not doCopy Anchor Link

It draws the guides the filters are listing, which is not necessarily every guide the nuclease could use — a track that is empty in a region may mean no PAM, or may mean your thresholds excluded what is there. It does not evaluate the edit, model repair outcomes, or predict efficiency; the scores are columns in the table, and this is the geometry.