alignoth should display softclips with 50% opacity so they're distinguishable from regular matches.

The embedded view should contain a list widget where the user is able to add reads by clicking on them in the plot. When selecting the read in the list it should be highlighted in the plot. Of course there should be the option to remove reads from the list as well.

The html embedded plots should have a filter option that filters the alignment plots. This could be potentially solved with vega-lites own filter function so the filter ui will be generated automatically.

When using vl2pdf the current specs throw the following error: Error: CanvasRenderer is missing a valid canvas or context. This can be currently bypassed by using vl2vg | vg2pdf but should be fixed anyways.

Hello Team,

what could be the error here:
thread 'main' panicked at 'called Option::unwrap() on a None value', /storage/home/hcoda1/4/jzhao399/.cargo/registry/src/github.com-1ecc6299db9ec823/alignoth-0.6.0/src/plot.rs:30:58
note: run with RUST_BACKTRACE=1 environment variable to display a backtrace

Thanks,

Jianshu

Currently paired reads are always placed in the same row in the plot even if they overlap. If they do overlap the mate should be placed in the row below and if possible with vega-lite connected via a thin zig zag line.

The tooltip should contain information from the XA tag under the title alternative alignments.

This happens because trailing_softclips() returns 0 if there are hard clips that follow e.g. 50M5S10H. This would be fixed by rust-bio/rust-htslib#375.

Using the Vega-lite selection functionality we should be able to easily allow people to select a single read pair and make all the other ones more translucent: https://vega.github.io/vega-lite/docs/selection.html#project

This would be nice for highlighting individual items e.g. in a presentation situation.

This should allow omitting the region in the command line call and should only be used for small bams. The region to be plotted can be retrieved from the first and last read of the bam file. If the file contains reads from multiple chroms or > 100000 reads an error should occur.

There seems to be a problem when using the subsample option:

thread 'main' panicked at 'called `Option::unwrap()` on a `None` value', src/plot.rs:352:58
note: run with `RUST_BACKTRACE=1` environment variable to display a backtrace

For easier integration and usage with datavzrd, alignoth should be able to output read, reference and highlight data in tsv format. This should be solved by adding a new parameter to the cli interface: --data-format [json|tsv].

The highlight parameter should also accept a single position to highlight a single base position in the plot.

The new --around parameter should allow the user to generate a plot around the given base position +-500bp.

Currently alignoth simply unwraps on a none value in the following line: https://github.com/koesterlab/alignoth/blob/32e5decdca43414bd6ae343b9bd24a33e274f7ea/src/plot.rs#L30

If the directory from which alignoth is called only contains one file of each the reference and bam arguments shouldn't be necessary.

The tooltip of deletions should include its length.

Hello,

I'm quite interested in your project but I think it would be nice to have an idea of the output plots in the README.

Have a nice day :)

It would be fantastic if Alignoth could also display other annotation tracks, such as features from a GFF or BED file.

I'm not sure if this is out of scope for your vision of the tool, but it seems like a promising replacement for clunky IGV batch screenshots.