Selection / Colouring of Observational Data table

[Don-Isdale]

Some notes :

colouring base values (A/C/G/T) in the table cells

• A Green
• C Blue
• G Red
• T Black

Alternate mode, when an individual column is selected, display instead of A/C/G/T : A/B, where

• A means == the value of the selected individual,
• B means !=

conventional colours : A : green, B : red (terminology: A "reference value", B "alternate value")

See also : https://ics.hutton.ac.uk/flapjack/flapjack-screenshots/

hover on a cell : highlight that feature on the aligned genetic maps

We can have an invisible axis with constant height feature spacing - lines up with the rows of the table, so the paths appear to align to the left edge of the table, or the table can report the position of each row { "myFeature2" : yPixel_offset } and the paths can be drawn to these values.

(later) These lines, if too crowded, may be spread out like a Sankey Diagram : https://github.com/d3/d3-sankey https://github.com/d3/d3-hierarchy/blob/master/README.md#tree

(later) Relationship between individuals may be shown as a tree, e.g. https://bl.ocks.org/mbostock/c034d66572fd6bd6815a Tree of Life https://www.researchgate.net/figure/Maximum-likelihood-ML-tree-inferred-from-D2-D3-28S-LSU-rDNA-for-several-species-of-the_fig9_323965407 https://bl.ocks.org/mbostock/4339083 Collapsible Tree

[Don-Isdale]

For the purpose of classifying possible features to be added to observational table, I'm collating a check-list; seeking comment on whether features are relevant to Pretzel or not, and a priority / order.

We want to select the essential features for June, others can be started after those are released, but it is useful to assign a likelihood to later features, so we can make sure the UI design is integral.

already mentioned in the #dev comments :

• aligning the table with mapview; (put the matrix in the right panel) The left panels View/Explorer/Upload - may operate in 2 separate modes for mapview / individual, or can they each contain 2 contexts, perhaps separate tabs mapview & individual. (Alex, any thoughts ... )

• clicking on the column heading sorts, by data value in that column, and re-arranges other columns in parallel; does this apply to observational table ?

• allow user to create additional tables and select which table to add individual / annotation to.

• possible additional means of selecting focus individual, e.g. click on column header

Additional :

• display any available metadata of selected row / column of table (individual / feature)

• extend individual focus from single to multiple , i.e. A if == all of the selected focus individuals, B if != all, and otherwise A/B with a transient colour between green & red.

some possible features :

• individuals sorted in alphabetic order by name
• scrollbars H & V on the main display
• cells with composite values e.g. A/G - cell divided into 2 triangles.
• find individuals (line names); options for matching case, regexp. (possibly auto-complete would be a good approach, within the individual names in the table) Maybe filtering is relevant - i.e. the matching individuals are displayed or selected.
• overview : provides a zoomed-out view, could allow brushing a rectangular region, which would then focus the main display on that region; brush can be dragged to pan.