Document why Base pair match lines on the sequence view don't always connect all the sequences

[diane]

in cases where there's a match between the first and second sequence but not the third, it will draw a line from the first to second even though there's no match to the third.

this is wrong, lines need to go all the way through only when there's a base pair match all the way through.

[diane]

Ok I wasn't able to duplicate the only 1 line segement error but I did notice that the lines connecting sequences are off. Currently it looks like all of the lines are being drawn about 1 unit further right than the tracks.

As for what this bug was originally about [409] seems to suggest that at least at the low level alg module it appears to work.

[diane]

Hmm... it might be that the base pair it was connecting to was way accross the screen?

I just saw one data file where some of the "aligned base pairs" were a long way away from most of the alignment. on my system ~/proj/mussa_analyses/myogenin/HuChMoXeZe/myog_w30_t21 seems to illustrate the problem.

[diane]

Ok I have a guess at what the actual problem is now, and I think mussa is acting correctly.

You can get the illusion of a base pair match that stops in a multi-way comparison when there's an overlapping set of windows and one of the windows that is being matched to is far off the screen. The lines leading far off the screen all blend together and it just looks like everything stops.

Thus this is now a documentation problem as mussa appears to be correct, its just really confusing.

[diane]

See OverlappingWindows

[king]

Done in change set [465].