charlesbreeze/eFORGE

Rename bkgdstats and related variables for cleaner code

Even more code cleanup

More code cleanup

Clean up the code

Fix mixup between strict and marginal thresholds of significance. Als… …

…o fixes pink color on dChart

Fix buglet: PDF was always showing default thresholds in the legend

Use SQL query compatible with old RDBM

New DB schema - both code and DB are data-agnostic

Copyright and authorship

Move SQL for fetching probe_id from location to eForge.pm

Fix bug about remaining variable

Fix legacy from FORGE in process_file and fetch_rsid (now called fetc… …

…h_probe_id)

Support to deploy the webserver on any URL (like http://server.org/ef… …

…orge/)

Support for cells from diff tissues with the same name …

The plot support cases where two tissues have cell type with the same name. This happens in the
Blueprint data set, for instance.

changes to email and copyright (still GPL) …

contact and added UCL copyright details (still GPL)

db changes in code …

changes in database to add blueprint and consolidated roadmap data

Increase fonts and contrast for readability

Increase fonts, parametrise dPlot size, reorder objects

Clean up code + Fix Bonferroni-by-tissue correction

Adds title to the dPlot and change font to Arial

Keep p-value in natural scale and represent the -log10 in ePlot

Avoid using sed which behaves differently in Mac vs Linux

New vertical lines + labels in dPlot

Fix bug in zscore method

Add zscore method

Clean up the code …

- Use single quotes within double-quoted text
- Comment the R code
- Remove deprecated rPlot method
- Simplify the table method
- Other cosmetic fixes

Fix last cosmetic change

Cosmetic change to avoid any overlap between the first cell and the l… …

…egend

FDR correction + similar dPlot and PDF …

The points are now colored according to the q-value instead. Also, the PDF output has changed
to be more similar to the dimple plot. The dimple plot now is also more similar to the PDF
as the cell lines/types are sorted in the same way and the break lines between tissues are
also shown.

Fix bug in sed command

Rename mvp_bins to mvp_450k_bins

Clean up the code after merge

27k hash addition …

Modification of eFORGE hash selection to include 27k array as an option
for background (subset of 450k probes that are on the 27k array)

Add support for whitespaces in output dir

Fix buglet for BED input …

The code was using the end coordinate instead of the start one when looking for the probe in the DB.