Skip to content

Latest commit

 

History

History
112 lines (89 loc) · 3.09 KB

release-notes.md

File metadata and controls

112 lines (89 loc) · 3.09 KB

===2101===

  • initial version

===2102===

  • patterns: added >30 SMARTS, replacing unsuitable wildcard SMILES
  • biosyn: added 24 SMARTS
  • classify: add option to load superclass dataset; does nothing atm
  • add class-subclass.py for updating superclass dataset

===2103===

  • unified dataset handling (all datasets are loaded from given directory)
  • data update

===2104===

  • classify: remove redundant hits
  • classify: get nearest parent GO process if none on class
  • classify: add option to get redundant hits as before
  • superclass data: more class types, bigger ontology, dead-end diagnostic

===2105===

  • all data: monoterpenoids
  • test data: monoterpenoids, 14/197 FAIL
  • classify: test capability
  • multiple SMARTS (denoting sets of patterns) now handled

===2106===

  • all data: sesquiterpenoids
  • test data: sesquiterpenoids, 7/240 FAIL

===2107===

  • all data: diterpenoids
  • test data: diterpenoids, 0/177 FAIL
  • don't load SMILES/SMARTS if InChI present, 10x speed up of classification

===2108===

  • remove bogus pattern leading to Boost warnings
  • revised/added to acyclic monoterpenoids, FAIL: 0/197
  • all data: sesterterpenoids
  • test data: sesterterpenoids, 1/14 FAIL
  • split pattern query, makes it faster, more resilient, one third reduction in pattern db size

===2109===

  • some new files were missing in last release
  • reduce test run output
  • all data: triterpenoids
  • test data: triterpenoids, 1/85 FAIL

===2110===

  • pattern fixes (not all possible double bonds were allowed)
  • all data: tetraterpenoids
  • test data: tetraterpenoids, 7/199 FAIL

===2111===

  • all data: steroids
  • test data: steroids, 3/1561 FAIL

===2112===

  • add -n option to classify small sample of WD natural products
  • all data: flavonoids pt. 1 (est. 60% of all)
  • test data: flavonoids

===2113===

  • all data: flavonoids pt. 2 (est. 85% of all)

===2114===

  • add -j option for JSON output (NP only)
  • all data: flavonoids pt. 3 (est. 95% of all)

==2115==

  • all data: alkaloids pt. 1 (est. 25% of all)
  • test data: alkaloids

==2116==

  • fixed bug where compounds with biosyn process were not seen with -j option
  • all data: alkaloids pt. 2 (est. 45% of all)

==2117==

  • sort pattern data on SMARTS hash to avoid most reshuffling
  • all data: alkaloids pt. 3 (est. 62% of all)

==2118==

  • fix bug where hits to NP root classes were ignored
  • again fix data write order; Py hash() was salted
  • add rule to recognize unspecified alkaloids
  • all data: alkaloids pt. 4 (est. 35% unspecified)
  • add option ouputting NP roots in JSON format
  • revamp top NP hierarchy

==2119==

  • all data: polyketides pt. 1 (est. 90% of all)
  • test data: polyketides
  • prettify --list_nproots
  • further NP root optimizations

==2120==

  • test data polyketides was missing
  • add rule to recognize unspecified macrolides
  • all data: polyketides pt. 2 (est. 97% of all)

==2121==

  • classify: revamp -n code
  • classify: include mol data with JSON output
  • improve macrolide rule
  • reduced unspec. alkaloids to 25% by adding more small groups
  • unspec. alkaloids and makrolides tagged in WD

==2122==

  • all data: fatty acyl pt. 1 (est. 77% of all)
  • test data: fatty acyl
  • classify: fatty acyl rules