meganap / multivis Goto Github PK

be able to add a space between sample sets so it's easier to see where one set of samples ends and another begins

sort by abundance of some taxa (e.g. abundance of Firmicutes) and be able to do so within a mapping field category (e.g. within country sort by increasing abundance of firmicutes). Looks like Meg's multivis comes close to doing this if you can Sort by abundance of a taxa as well

As the title indicates

Add feature to drawing function so that drawing labels depends on size of bars in tax plots

Need to do some code profiling or use some tricks to make it easier to interact with the bar charts when there are on the order of 10k samples

Currently multivis sources javascript from multiple places online which makes its use dependent on a network connection

be able to control how much info goes into the legend labels (e.g. if using level 6 taxonomy, perhaps you just want to list family and genus and then have bars colored by L2 taxonomy - see above suggestion about coloring). hide part of the info. Definitely get rid of the “p__”

be able to filter out taxa below some minimum threshold

Ability to drag stacked bars in stacked bar view

for the finer levels of taxonomic identification (below family), be able to see the % similarity to the reference

Color by metadata functionality
Change color of particular item

Color themes to choose from
-continuous/gradient coloring
-discreet coloring
-hierarchical coloring

Support dynamic coloring by other observation metadata besides taxonomy, if present in the biom table (e.g. predicted or measured organism lifestyle or rRNA copy number). Would allow plotting for example predicted heterotrophs vs. phototrophs or aerobes vs. anaerobes.

Compute is sluggish, but nothing is shown on this table

If you change the sort_by value without changing the group_by value, the X axis labels will disappear.

When loading the index in Firefox, the bar and donut graphs are completely black and throw the following error:

Error in parsing value for 'fill'. Declaration dropped.
Expected color but found 'ff4000'. Error in parsing value for 'fill'. Declaration dropped.

While BIOM does support metadata internal for samples, we need to support passing a metadata file in that is separate from BIOM

@ElDeveloper and @antgonza, can you comment on the current Javascript style in use and if you have any recommendations on stylistic changes, guidelines, standards, etc?

be able to click on a taxa and test whether it's abundance is significantly different across some category and make an accompanying histogram with error bars (e.g. Firmicutes abundance by age). Maybe this is similar to the "Generate OTU significance tables" in Justine's talk

Implement area charts

Add BSD license