LimROTS: A New Tool for Reliable Differential Expression Analysis in Proteomics
Differential expression analysis is central to modern omics research, helping scientists identify proteins, genes, or other molecular features associated with different biological conditions. In proteomics, these analyses can be complicated by technical variation, batch effects, small sample sizes, and biological covariates such as age and sex [1] [2].
To address these challenges, our team developed LimROTS, an R/Bioconductor package that combines two complementary statistical approaches: the flexible linear modelling and empirical Bayes framework used by limma, and the reproducibility-optimized statistical approach implemented in ROTS [1] [2] [3].
The limma framework uses linear models to accommodate complex experimental designs and empirical Bayes methods to improve statistical inference by borrowing information across many molecular features [2]. ROTS, in contrast, optimizes a ranking statistic according to the reproducibility of results across bootstrap samples, allowing the statistic to adapt to the characteristics of a dataset [3] [4].
LimROTS integrates these ideas into a moderated ranking statistic. This enables reproducibility-optimized differential expression analysis while allowing technical and biological covariates to be incorporated into the statistical model [1].
The method was validated using 21 gold-standard proteomics spike-in datasets representing different protein mixtures, mass-spectrometry instruments, and experimental techniques. Its performance was compared with commonly used approaches including limma, ROTS, MSstats, DEqMS, DEP, SAM, ANOVA, and t-tests [1].
Across diverse benchmarking settings, LimROTS showed strong performance and, in the published evaluation, outperformed comparison differential-expression methods across several important measures and complex experimental contexts [1].
We also evaluated LimROTS using clinical Alzheimer’s disease proteomics data. This case study illustrated how the method can be applied to biologically complex datasets where technical and clinical covariates are important considerations [1].
LimROTS is openly available as an R/Bioconductor package and supports standard Bioconductor data structures such as SummarizedExperiment, helping it integrate with existing omics analysis workflows [1] [5].
The method was developed and benchmarked primarily for proteomics, while the authors note potential relevance to other high-dimensional omics domains such as transcriptomics and metabolomics, subject to further validation [1].
Explore LimROTS at: https://bioconductor.org/packages/LimROTS
References
[1] Anwar, A. M., Jeba, A., Lahti, L., & Coffey, E. (2025). LimROTS: a hybrid method integrating empirical Bayes and reproducibility-optimized statistics for robust differential expression analysis. Bioinformatics, 41(12), btaf570. https://doi.org/10.1093/bioinformatics/btaf570
[2] Ritchie, M. E., Phipson, B., Wu, D., Hu, Y., Law, C. W., Shi, W., & Smyth, G. K. (2015). limma powers differential expression analyses for RNA-sequencing and microarray studies. Nucleic Acids Research, 43(7), e47. https://doi.org/10.1093/nar/gkv007
[3] Suomi, T., Seyednasrollah, F., Jaakkola, M. K., Faux, T., & Elo, L. L. (2017). ROTS: An R package for reproducibility-optimized statistical testing. PLOS Computational Biology, 13(5), e1005562. https://doi.org/10.1371/journal.pcbi.1005562
[4] Elo, L. L., Filén, S., Lahesmaa, R., & Aittokallio, T. (2008). Reproducibility-optimized test statistic for ranking genes in microarray studies. IEEE/ACM Transactions on Computational Biology and Bioinformatics, 5(3), 423–431. https://doi.org/10.1109/TCBB.2007.1078
[5] Johnson, E. C. B., Dammer, E. B., Duong, D. M., et al. (2020). Large-scale proteomic analysis of Alzheimer’s disease brain and cerebrospinal fluid reveals early changes in energy metabolism associated with microglia and astrocyte activation. Nature Medicine, 26, 769–780. https://doi.org/10.1038/s41591-020-0815-6
[6] Bioconductor. LimROTS: A Hybrid Method Integrating Empirical Bayes and Reproducibility-Optimized Statistics for Robust Differential Expression Analysis. Bioconductor package. https://bioconductor.org/packages/LimROTS