-------------------
* `-l` and `-s` CLI arguments and corresponding `length` and `stride` parameters
to `Goldilocks` constructor now support SI suffixes: `K`, `M`, `G`, `T`.
`util` module contains `parse_si_bp` used to parse option strings and return
the number of bases for length and stride.
* Add length and stride to x-axis label of plots.
* Add `ignore_query` option to `plot` to override new default behaviour of plot
that only plots points for regions remaining after a call to `query`.
* Remove `profile` function, use `plot` with `bins=N` instead.
* Add binning to `plot` to reduce code duplication.
* Add `chrom` kwarg to `plot` to allow plotting of a single chromosome across
multiple input genomes.
* Fix support for plotting data from multiple contigs or chromosomes of a single
input genome when provided as a FASTA.
* Add `ignore_query` kwarg to `plot` for ignoring the results of a query on
the `Goldilocks` object when performing a plot afterwards.
* Bins no longer have to be specified manually, use `bins=N`, this will create
N+1 bins (a special 0 bin is reserved) between 0 and the largest observed
value unless `bin_max` is also provided.
* Bins may have a hard upper limit set with `bin_max`. This will override the
default of the largest observed value regardless of whether `bin_max` is smaller.
* Plots can now be plotted proportionally with `prop=True`.
* Improve labels for plotting.
* Reduce duplication of plotting code inside `plot`.
* Share Y axis across plot panels to prevent potentially misleading default plots.
* Reduce duplication of code used for outputting metadata:
* Add `fmt` kwarg to `export_meta` that permits one of:
* bed
BED format (compulsory fields only)
* circos
A format compatible with the circos plotting tool
* melt
A format that will suit import to an R dataframe without the need
for additional munging with reshape2
* table
A plain tabular format that will suit for quick outputs with
some munging
* Remove `print_melt`, use `export_meta` with `fmt=melt`.
* Add `is_pos_file` kwarg to Goldilocks, allows user to specify position based
variants in the format `CHR\tPOS` or `CHR:POS` in a newline delimited file.
* Changed required `idx` key to `file` in sequence dictionaries.
* Added custom strategy and plotting examples to the documentation.
* The `Goldilocks` class is now imported as `from goldilocks import Goldilocks`.
* The `textwrap.wrap` function is used to write out FASTA more cleanly.
* A serious regression in the parsing of FASTA files introduced by v0.0.80 has
been closed.
* Improved plotting functionality for co-plotting groups, tracks of chromosome
has been introduced. Tracks can now be plotted together on the same panel by
providing their names as a list to the `tracks` keyword.
* `reset_candidates` allows users to "reset" the Goldilocks object after a
query or sort has been performed on the regions.