June 15th: Refactor percentiles to be lab specific for all items
May 17th: Released revised UI
Revisions are still ongoing, but sufficiently stable
May 12th: Updated Long COVID publications and Bacteria
April 23, Release Probiotics suggestions based on symptom aggregation and enzymes for each sample which has enzymes. Often this reduces the enzymes to increase by 75% and should be superior to the generic probiotics suggestions released below.
April 22, Added Probiotics for Symptoms based on KEGG Enzymes (not sample specific – generic), see this post
Apr 4-21 Implemented Categoric Logistic Regression to Symptoms for:
Compounds estimated from KEGG
Enzymes estimated from KEGG
April 3, Refactor KEGG Probiotic Suggestions, see post.
April 2, Added/Update Bacteria Healthy ranges on per bacteria basis. The computed ranges are linked to pages explaining the method
April 1, Implement periodic update from NCBI on new bacteria (on 21st of each month). Today, there was 2,419,768 different bacteria classified and assigned NCBI numbers
Mar 27,investigate odd Kaltoft-Moldrup ranges. Discovered typo in formula, running tests and recalculations.
Mar 26, Change all distribution charts to real time (Taxon broke for a short while — now fixed)
Feb 12, implemented Display Levels to allow people to better handle information overload.
February — rework the KEGG data to estimate species from existing strain information
Jan 19, Added links to Bacteria Details from Researched Probiotics Page
Jan 17, added Kegg Detail pages (for bacteria in KEGG)
Jan 15, Fixed bacteria selection bug on GMRepro page
January – onwards. Move periodic updates to a dedicated machine (16 GB memory, SSD Drive, 8 processors) and proceeding to replace old adhoc code with consistent new code. as well as improve performance. (The first complete update run of the old code was over 72 hrs, this has been greatly reduced). Work is still ongoing (mainly catching issues)