matt-sd-watson / spora Goto Github PK

Need to fix logic for using the --names-csv to rename FASTA headers and read in the CSV to match new headers to the old ones based on the focal input.

maybe using phylocanvas or another Bioconductor library supported by R markdown?
https://github.com/zachcp/phylocanvas

Use of --rename and --names-csv in the output report causes an error in the detection of the focal and background names for SNP distance.

In the draft version of the summary report, the txt file or multi-FASTA for both the focal and background sequences are read in for summary statistics. If any of the sample names in the tree match to the focal sequence list, they will be categorized as "focal", or "background" otherwise.

Currently this breaks if --rename is used, as the focal and background names are read from the files before the renaming, so a mismatch will occur and no focal sequences will be detected.

--rename with --names-csv requires that all sequences supplied in the CSV exist in the multi-FASTA. if filtering removes any sequences, this breaks the fastafurious rename command.

Likely need to change the order of rule filter and rule rename to be compatible with the removal of sequences.

Have the summary report be able to figure out where to render the geom_treescale() using ggtree in the trees depending on branch locations (i.e. do not have the scale placed over a branch).

for compatibility with PyPi

Maybe make the size relative to the number of labels in the tree, and cap a minimum.
Can be used to more easily view the labels for trees with fewer samples.

Need to be able to have the labels for trees in the summary report to automatically be visible within the margin of ggtree. Will require scaling with xlim when rendering the tree based on the maximum branch length.