Filtering based on samples in same condition

[hsiaoyi0504]

Currently, we only do filtering for LFQ data in the process. However, TMT and DIA also need some kind of filtering but probably not based on replicates only. For example, we don't have many replicates in CPTAC TMT data. Without filtering data before imputation is not pretty ideal.

[hsiaoyi0504]

We could have two options listed in configuration:

• filter by replicates
• filter by condition

By default, LFQ will use filter by replicates, but for TMT and DIA, we will use filter by condition.

[hsiaoyi0504]

just two filters

• min frequency X and
• min frequency X in at least 1 condition

if I want just 50% across the whole table, it will be 100, 0 if per one condition 0 100 default 0 0 meaning no filter

• user case 1: X% across the whole dataset (50% as in CPTAC)
• user case 2: min X% in at least one condition

[hsiaoyi0504]

Finished in 19f1c072e6aee84f4a2caa3c8a478cf7887d9e1b.