New bootstrap_ci() function can be used to compute a bootstrapped confidence interval for COI estimates (#24, #25).
An autoplot and plot method have been written for simulated data (#22).
Simulated data have been restructured and are now assigned a class sim (#22).
The default sequence error threshold has been set to 1%.
Our methods now can estimate the COI without grouping data points into buckets (#17).
Within sample minor allele frequencies are now weighed by read coverage. If no coverage is supplied, the coverage is assumed to be uniform across all loci (#16.)
New theme_coiaf() creates a custom theme for this package.
Bug Fixes
Midpoint calculation for buckets has been improved. Previously, if buckets consisted of a single value, we would combine the bucket with another bucket. We no longer do so and let buckets consist of only a single value (#19).
In our estimation methods, each bucket is now weighed by the number of points in each bucket (#15).
The Frequency Method is undefined for a COI less than two. When we suspect the COI should be less than two, we return NaN. We additionally return a note indicating that the COI may be one. If we are able to compute the COI, we also return the COI estimated by our methods (#14, #21, #23).
Maintenance
Fix broken links.
Deprecate arguments in favor of newer methods to estimate the COI. Arguments deprecated include: comparison, distance, use_bins, and bin_size (#26).
PLAF has been renamed to PLMAF to indicate the focus is the minor allele (#13).
WSAF has been renamed to WSMAF to indicate the focus is the minor allele (#13).
COI methods have been renamed for improved clarity. "Method 1" has been renamed to "Variant Method" and "Method 2" has been renamed to "Frequency Method" (#9).