3.0.1 - GA memory footprint limiter

Added a new dynamic job limiter for GA:

• removes the need to calibrate concurrent instances of BWA/BLAT/DMD
• automatically restrains the number of concurrent GA instances based on footprint and total system memory to prevent crash.
• new option in config[Settings]: BWA/BLAT/DMD_mem_footprint

3.0.0: GA-pre-scan

Changes:

replaced kaiju with kraken2
new GA-pre-scan stage
putative reads will now be scanned by centrifuge + kraken2 to determine likely taxa in sample
MetaPro will now craft a custom subset of ChocoPhlAn to speed up GA (BWA+BLAT) steps
ability to rename stages
ability to change the name of the bypass log
debug_stop_flag to stop MetaPro at a specific stage
Bugfixes:

output_per_read_score now writes to bypass log

2.1.1: DETECT temp naming issue

critical bugfix:
-> DETECT temp export name too long.
-> new name handler. (and unique header-ID-maker)
-> Handles choco h3 names

minor bugfix:
-> removed artifact from default Java path in MetaPro_paths.py
-> removed unused paths in MetaPro_paths.py

2.1.0: PRIAM-changes

changes:
-> new protein split function for PRIAM throughput
-> -> split proteins
-> -> re-tooled PRIAM launcher for split-proteins
-> -> new merge step for PRIAM
-> -> rerouted ec_post to take in merged PRIAM output

-> patched the EC_priam time reporting
-> patched the "missing BLAT_pp_v2.py". now points to v3
-> removed --cg in PRIAM arguments

2.0.5: hotswap of scripts.

new features:
-> ability to adjust internal scripts with the config

changes:
-> refactored metapro_paths (cleanup + hotswapping script versions)
-> new ga_blat_generic _v3.py that doesn't rely on seqIO.index

assembly_bypass patch

Fixed bug that caused the pipeline to not write "assemble_contigs" in the bypass log on successful run

2.0.3: single-file DB handling

Added checks for single-file Database.
Added a bypasser for GA_BWA_merge so that single-file DB skips this (it's not used. nothing to merge).

2.0.2

Major:
-> streamlined top-level to be more modular / easier to insert new stages
-> re-using new modules for single-shot + tutorial mode

Minor:
-> cleaned up DNA_db line so there's only 1
-> added DB status checks

bugfixes:
-> wait issue with GA final merge proteins
-> support for DNA_DB path optionally ending in '/'
-> MGM license check
-> bypasser for contig formation in small datasets.