Clickable urls in global n local feature table
Italise species names in global and local feature table

Examples

• ttf_Maximum_of_free_energy_ratio_of_miRNA:mRNA_duplex
• ttf_Number_of_differentially_methylated_CHG_in_gene_body_of_TG_(Kawakatsu_et_al.,_2016)

Red and blue edges could be too glaring. Change to a lighter shade

Stuff to manually check in google storage:
agi_txt/gbm_Gene body methylated -> models, scores, fi files, need to check to make sure there's no spaces
(after their respective files are updated, pls check again)

Misc:
cin_ATB2/AtbZIP53/AtbZIP44/GBF5 BS in ProDH -> has extra space at the end, in original ml dataset, messes up database, solved by adding space at the end of this, in the local neighbours network python script

Modified error bar to explain why GO terms with <10 genes can't be searched, as machine learning isn't trained on them -> Display a better result? Maybe just some general info without network, feature importance info and frs scores?

Create readme files and reorganise the following files for download:

• models
• feature importance files
• scores

gzip all these files, so that there's 3 gzip files only, can combine all these file which are of the same type together, don't need to separate into GO, DGE etc, like what I did previously.

Write readme files for the above, and put them in logical places in the folder directory

Clean up code so thats its easier to read e.g.

• network code (edge parameters etc) in network.js
• goMethods.py code

Change colour, as stated in the title

Recent updates this week to markupsafe and itdangerous broke the database, solved by fixing the version and preventing them from upgrading to newer versions. Used the below in requirements.txt.

markupsafe==2.0.1
itsdangerous==2.0.1

References
pallets/markupsafe#284
https://serverfault.com/questions/1094062/from-itsdangerous-import-json-as-json-importerror-cannot-import-name-json-fr