Table 1 Examples of diverging decisions between imputation methods

From: Imputation of label-free quantitative mass spectrometry-based proteomics data using self-supervised deep learning

Protein Group

F5H8B0

P08709

P08709-2

A0A0D9SG88

A0A075B6R2

P02741

I3L0A1

J3KPA1

P54108

P54108-2

P54108-3

P59665

P59666

Gene

F7

CFH

IGHV4-4

CRP

CRISP3

DEFA1B

DEFA3

None (ref.)

0.954

0.117

0.000

0.030

0.018

0.365

DAE

0.916

0.022

0.000

0.038

0.001

0.025

TRKNN

0.470

0.059

0.000

0.029

0.000

0.082

VAE

0.918

0.038

0.000

0.039

0.000

0.023

RF

0.915

0.061

0.000

0.040

0.000

0.061

CF

0.841

0.079

0.000

0.031

0.001

0.022

Median

0.937

0.113

0.000

0.030

0.051

0.373

RSN

0.082

0.951

0.080

0.052

0.745

0.012

QRILC

0.002

0.727

0.003

0.062

0.626

0.029

  1. Based on comparison of q-values, i.e. multiple testing corrected p-values using the Benjamini-Hochberg method, for previously included protein groups. Analysis without imputed values is denoted as None and used as reference (ref.). All diverging decisions are in Supplementary Data 11,12.