obi-ontology / obi

The Ontology for Biomedical Investigations
http://obi-ontology.org
Creative Commons Attribution 4.0 International
74 stars 24 forks source link

_unclassifed in dt branch - move multiple testing terms #212

Closed obi-bot closed 7 years ago

obi-bot commented 14 years ago

proposal to add objectives for the following and remove from unclassified:

1. add new objective multiple testing correction objective as child of correction objective. Note def of correction objective implies only data error correction and this needs to be modified to deal with multiple testing which is not data related AFAIK

apply to logical defs of, also add these as child classes of multiple testing correction method. Rename this class to multiple testing correction - this is a process not a method.

Benjamini and Hochberg false discovery rate correction method Benjamini and Yekutieli false discovery rate correction method Holm false discovery rate correction method Westfall and Young family wise error rate correction

2. add model validation objective (or statistical model validation objective) apply to

k fold cross validation leave one out cross validation

3. boostrapping and jacknifing are both resampling techniques used for model validation (and other purposes which can be added later)

2. remove multiple comparison ajdustment - as multiple testing correction method already present

4. remove data pooling from _unclassified and add logical def 'merging objective'

Reported by: helenp

Original Ticket: obi/obi-terms/215

obi-bot commented 14 years ago

5. longitundinal data analysis ->change to - longitudinal data analysis objective 6. survival analysis ->change to survival analysis objective 7. Kaplan Meier- has objective 'survival analysis objective' has objective 'longitudinal data analysis objective' - spec_inout only data, spec_output=only data 8. (mass) spectrometry analysis -. objective the child terms of mass spec analysis will have this objective 8a. continuum mass spectrum 8b. characteristic path length calculation 8c. centroid mass spectrun-> centroid mass spectrum

9. spectral integration with user-controlled baseline correction - need more info - has correction objective. Appears to apply to NMR spectroscopy data maybe make 8. general to also cover this e.g. http://mrs.cpmc.columbia.edu/hires.html 10. non negative matrix refactorization http://en.wikipedia.org/wiki/Non-negative\_matrix\_factorization

* consider using restricted input/outout data items when defining these e.g. error corrected data set

Original comment by: helenp

obi-bot commented 14 years ago

mutiple testing correction class made defined, existing children given axioms, terms from unclassified also given axioms removed from assertion under _unclassified, still needs some work on error correction data transformation, as this is now not different from multiple correction dt, which is a subclass - these are correctly equivalent, need to delete one and fix the child terms prolly.

Original comment by: helenp

obi-bot commented 14 years ago

Only mass spec terms and survival analysis left to deal with under _unclassified, Tina's working on defs.

Original comment by: helenp

obi-bot commented 14 years ago

Only mass spec terms and survival analysis left to deal with under _unclassified, Tina's working on defs.

Original comment by: helenp

obi-bot commented 14 years ago

Original comment by: helenp

obi-bot commented 14 years ago

Original comment by: helenp