To do

• Conserved log class
  • Must be capable of determining if the run is congruent or new
• Annotate code (all)
  • add type constraints to functions
• Conform old scripts to PEP8 (all)
• [o] Build all-in-one stable conda package
• Transition code-base to Rust
• implement kon_log class, paying attention to verbosity control
• kon_log class output to local configuration directory
• support compressed databases
• register mycotools with NCBI
• configuration to output mtdb name + date in output directory
• discuss multigene phylo pipeline
• discuss db2hgs
• uniform argument parser
• allow for inputting multiple DBs with full paths
• robust conda update checks

crap

• Outgroup manager for clusters that fit within min and maximum sequences
• Percent positives filter
• Integrate agglomerative clustering
• Allow for inputing a specific run order
• Log-based resume
• Do not reiterate running a gene in the same homology group
• Allow converting HG runs' names
• Better root inference
• Assembly query method, i.e. through tblastn
• Allow changing the clustering variable
• locus output using percent similarity

curGFF3

• pseudogenes can have RNAs, and CDSs from NCBI may reference those pseudogene parents or their RNAs (GCA_900074715.1_LAME0)
• some pseudogenes fail because they are given an "Alias" without being completed (GCA_004920355.1)
• make universal interface to remove need for source column
• allow including entries that cannot be hiearchically assimilated into genes or transcripts
• build to universally include genes and transcript-assimilated types

db2hgs

• implement db2search to identify NSCHGs best-hits
• implement an automated NSCHG extraction based on minimum gene #

db2microsyntree

• Allow log removal

db2search

• Distinguish between nt and aa mmseqs dbs
• Allow for blastdb construction
• Streamline mmseqs parsing
• mmseqs save db option
• profile mmseqs search
• concatenate mmseqs query dbs
• optional fail upon any failures
• Log hmmer runs
• nhmmer option
• create all outputs as temp files and move when complete
• extract covered portion of hits
• max hits post blast compilation
• hsp option

dbtools

• Vectorize MTDB class
• make mtdb compiled class
• remove Entrez email login, simplify API access
• Get taxonomy of non-genus names
• get taxonomy XML - if it exists - instead of independent queries

extract_mtdb

• allow for a lineage list from command line (may already be integrated)
• stdin argument input
• Fix when lineages have multiple ranks, e.g. Tremellales sp. will be extracted from Tremellales input, when the order is likely what's requested

fa2clus

• sort log by default, and only unique run parameters
• percent positive mode
• integrate MCL
• rerun aggclus on new data

fa2hmmer2fa

• Move from extracthmm to simplified output parsing

fa2tree

• Implement fa2clus
• ignore non-fasta inputs
• take to phylogenomic tree from db2hgs

gff2svg

• find a prettier way to create SVGs
• parse for in gene coordinates and annotations
• create a single file output option for multiple inputs

jgiDwnld

• remove gff v gff3 option

manage_mtdb

• delete database feature
• fix local password encryption
• overwrite old password
• move database feature
• archive and unarchive genomes
• remove logfiles as parting of clearing the cache
• add combine DB option

mtdb

• add a log option of connected MTDBs
• remove standalone scripts from PATH
• look for old ome versions in query
• add a version querying option
• add an option to query log of analyses

ncbiDwnld

• db check to ensure log is relevant to input
• convert downloading to NCBI datasets
• add strain parsing from within GenBank records for entries that don't have an obvious strain entry

predb2mtdb

• source to reference the annotation source/project name
• integrate prokka/bakta
• error check FAA
• allow for just assembly accession in known sources
• allow inputting GBK

update_mtdb

• optimize dereplication, currently too slow
• initial JGI predb2mtdb fails because assemblyPath doesn't exist as a column, but restarts are fine
• allow updating from Predb.tsv immediately
• update introduction output
• need a verbose option
• reversion option
• reference a manually curated duplicate check
• prohibit specific IDs implementation
• finish --save
• singular strain download option
• pull failed JGI downloads from NCBI
• remove overlap when rerunning failed genomes
• central MTDB repository and reference option
• Improve MD5 check efficiency (update_mtdb)
• print organism name with genome accession
• don't remove files until after predb2mtdb (requires update_mtdb specific function)
• Need a manually curated file to correct errors in naming, e.g. Vararia v Vavraia, Fibularhizoctonia v Fibulorhizoctonia
• initialize from a predb
• option to remove entries that have been removed from genbank
• option to not dereplicate by genus and species alone
• main MTDB files for prokaryotes and fungi uploaded and that can be parsed
• add option to update taxonomy of existing entries
• sp. will also not dereplicate
• make add option check for overwriting entries (indicating incorrect PREDB linkage)
• ensure -t overlooks non-JGI/NCBI sources
• ensure assembly accessions from non-JGI/NCBI sources are not included in download
• still use a redundancy check when --ncbi_only is specified, or prevent changing between NCBI only and non-ncbi database
• acquire strain metadata and submitter organization from ncbi datasets