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
Currently, when evaluating a metric for scoring, it is possible to specify conditions that would make Mikado assign to it a score of 0. For example, if we are considering the metric non_verified_introns_num, we cannot explicitly say that this metric should be ignored when the transcript is monoexonic.
A solution could be allowing expressions of the form:
An issue has emerged: currently, when a transcript is excluded from the scoring for a metric because it runs afoul of the filter, its metric value is still being considered for calculating the boundaries of the scoring.
So for example if e.g. the scoring is on "maximum length of the 3'UTR, with a filter after 2500bps", if a transcript with a UTR of 3000bps is present in the locus no transcript will get the full scoring, because the maximum is 3000, not the longest UTR which does not run afoul of the filter.
This requires to rearrange the order of the operations within the scoring function.
Currently, when evaluating a metric for scoring, it is possible to specify conditions that would make Mikado assign to it a score of 0. For example, if we are considering the metric non_verified_introns_num, we cannot explicitly say that this metric should be ignored when the transcript is monoexonic.
A solution could be allowing expressions of the form:
This would allow to filter based on different metrics than the one examined.
The text was updated successfully, but these errors were encountered: