You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
A thing that caught me by surprise: now, if you simulate a chunk of a contig, you'll get 'missing data' outside that contig. This is represented as, basically, "no trees there". However, this will have ts.first().num_roots == ts.num_samples (and this is the reason for the root_threshold argument). This is fine and expected, if confusing - so maybe needs some documentation.
A particular gotcha is that if someone notices the ts doesn't look recapitated (although it is!), and recapitates again with pyslim.recapitate(ts, ...), then this will wrongly simulate coalescent history on the bits of chromosome that should be missing. I'm not sure if there's anything we can do to prevent this besides some documentation.
The text was updated successfully, but these errors were encountered:
A thing that caught me by surprise: now, if you simulate a chunk of a contig, you'll get 'missing data' outside that contig. This is represented as, basically, "no trees there". However, this will have
ts.first().num_roots == ts.num_samples
(and this is the reason for theroot_threshold
argument). This is fine and expected, if confusing - so maybe needs some documentation.A particular gotcha is that if someone notices the ts doesn't look recapitated (although it is!), and recapitates again with
pyslim.recapitate(ts, ...)
, then this will wrongly simulate coalescent history on the bits of chromosome that should be missing. I'm not sure if there's anything we can do to prevent this besides some documentation.The text was updated successfully, but these errors were encountered: