Table 1. Comparison of the Nine Evaluated Packages and Tools in Different Metrics and Properties^a.

	Open source	It has been benchmarked	QC reports	Input intensity data type	Normalization method(s)	Imputation method(s)
MSstats	√	√	√	Pep	Equalize median, Quantile, Global standards	Accelerated failure time model, cutoffCensore, censoredInt, MaxQuantileforCensored
msqrob2	√	√	×	Pep	quantiles, quantiles.robust, vsn, center.median, center.mean, max, sum, div.mean, div.median, diff.median	×
ProStaR	√	√	×	Pep/Prot	Global Quantile alignment and Centering, Column sums, vsn, LOESS	Det quantile, slsa, KNN, fixed value
Proteus	√	√	×	Pep/Prot	Equalize median, Quantile	Mean-variance (by limma package)
prolfqua	√	√	√	Pep/Prot	robust scale	Group mean imputation
ProVision	√	×	×	Prot	×	Normal distribution like Perseus
LFQ-Analyst	√	√	√	Prot	×	Normal distribution like Perseus, KNN, bpca, QRILC, MLE, MinDet, MinProb, min, zero
Eatomics	√	×	√	Prot	VSN	Normal distribution like Perseus, KNN, MinDet, QRILC
Perseus	×	√	√	Prot	Z-score, remove batch effect, Un-Z-score	A down-shifted Gaussian distribution, constant, ImputeLCMD

^aThe input intensity data type refers to the intensity data supported by the package: (Pep) peptide intensity; (Prot) protein intensity.