Ratio-based quantitative multiomics profiling using universal reference materials empowers data integration

Abstract

Multiomics profiling is a powerful tool to characterize the same samples with complementary features orchestrating the genome, epigenome, transcriptome, proteome, and metabolome. However, the lack of ground truth hampers the objective assessment of and subsequent choice from a plethora of measurement and computational methods aiming to integrate diverse and often enigmatically incomparable omics datasets. Here we establish and characterize the first suites of publicly available multiomics reference materials of matched DNA, RNA, proteins, and metabolites derived from immortalized cell lines from a family quartet of parents and monozygotic twin daughters, providing built-in truth defined by family relationship and the central dogma. We demonstrate that the "ratio"-based omics profiling data, i.e., by scaling the absolute feature values of a study sample relative to those of a concurrently measured universal reference sample, were inherently much more reproducible and comparable across batches, labs, platforms, and omics types, thus empower the horizontal (within-omics) and vertical (cross-omics) data integration in multiomics studies. Our study identifies "absolute" feature quantitation as the root cause of irreproducibility in multiomics measurement and data integration, and urges a paradigm shift from "absolute" to "ratio"-based multiomics profiling with universal reference materials.

Correcting batch effects in large-scale multiomic studies using a reference-material-based ratio method

“…PCA based performance metric Signal-to-Noise Ratio (SNR) was then used to quantify differences between biological sample groups and variations in technical replicates. SNR measures the ability of distinguishing intrinsic biological differences among distinct sample groups (“signal”) from technical variations including batch effects of the same sample group (“noise”), as mentioned in the accompanying papers 36, 39 . Generally, a higher SNR value indicates higher discriminating power, vice versa.…”

“…According to our results, using two or three replicates of a common reference material in each batch and converting expression data to feature-wise ratio-based scaling profiles within each batch can play an important role in making expression levels more comparable and resistant to batch effects. As the Quartet multiomic reference materials and the corresponding reference datasets have been successfully developed in our accompanying work 36-39 , which represent the first suites of publicly available multiomic reference materials, we therefore recommend the use of the Quartet reference materials for monitoring and correcting batch effects.…”

“…To examine impact of batch factors in multiomic data, we used reference materials and datasets from the Quartet Project. Establishment of Epstein-Barr virus (EBV) transformed B-lymphoblastoid cell lines and multiomic reference materials, as well as implementation of high-throughput experiments were described in an accompanying paper by Zheng et al 39 . Briefly, multiomic reference materials (DNA, RNA, protein, and metabolite) were derived from immortalized Epstein-Barr Virus (EBV) infected B-lymphoblastoid cell lines from a four-member Chinese Quartet family including two monozygotic twin daughters (D5 and D6), and their father (F7) and mother (M8) with expected differences in multiomic profiles between any two members 36-38, 40 .…”

“…Reference materials were then distributed to multiple labs for generating multiomic profiling data. According to the Quartet Project study design, in each omics, 12 samples were used as a standard sample set, consisting of 12 tubes with each representing one of the triplicates of a reference sample group 39 . The high-throughput experiments were conducted concurrently for the 12 samples.…”

“…Here, as part of the Quartet Project for quality control and data integration of multiomic profiling, we comprehensively assess the performances of seven BECAs for mitigating the impact of batch effects in multiomic datasets, including transcriptomics, proteomics, and metabolomics data. We previously established the Quartet RNA 36 , protein 37 and metabolite 38 reference materials derived from the same B-lymphoblastoid cell lines from the four members of a monozygotic twin family 39 . Large number of multiomic datasets were generated from multiple labs, platforms, and protocols.…”