Batch effect detection and correction in RNA-seq data using machine-learning-based automated assessment of quality

Abstract

Background The constant evolving and development of next-generation sequencing techniques lead to high throughput data composed of datasets that include a large number of biological samples. Although a large number of samples are usually experimentally processed by batches, scientific publications are often elusive about this information, which can greatly impact the quality of the samples and confound further statistical analyzes. Because dedicated bioinformatics methods developed to detect unwanted sources of variance in the data can wrongly detect real biological signals, such methods could benefit from using a quality-aware approach. Results We recently developed statistical guidelines and a machine learning tool to automatically evaluate the quality of a next-generation-sequencing sample. We leveraged this quality assessment to detect and correct batch effects in 12 publicly available RNA-seq datasets with available batch information. We were able to distinguish batches by our quality score and used it to correct for some batch effects in sample clustering. Overall, the correction was evaluated as comparable to or better than the reference method that uses a priori knowledge of the batches (in 10 and 1 datasets of 12, respectively; total = 92%). When coupled to outlier removal, the correction was more often evaluated as better than the reference (comparable or better in 5 and 6 datasets of 12, respectively; total = 92%). Conclusions In this work, we show the capabilities of our software to detect batches in public RNA-seq datasets from differences in the predicted quality of their samples. We also use these insights to correct the batch effect and observe the relation of sample quality and batch effect. These observations reinforce our expectation that while batch effects do correlate with differences in quality, batch effects also arise from other artifacts and are more suitably corrected statistically in well-designed experiments.

Temporal progress of gene expression analysis with RNA-Seq data: A review on the relationship between computational methods

“…An additional point to observe in quality assessment are batch effects. Batch effects arise from differences between samples that are not rooted in the experimental design and can have various sources, spanning from different handlers or experiment locations to different batches of reagents and even biological artifacts such as growth location [44] . Various methods have been developed to detect and/or remove batch effects in genomics data, particularly RNA-seq data.…”

Principles of phosphoproteomics and applications in cancer research

“…This highlights the reactivity of phosphorylation activity and the requirement for robust sample preparation protocols to create reliable datasets. However, these issues also greatly impact other omics methodologies such as single-cell RNA-Seq, RNA-Seq and CHIP-Seq [127][128][129].…”

Perspective on the challenges and opportunities of accelerating drug discovery with artificial intelligence

“…Because biological systems are complex, drug discovery experiments often possess conditional features whose importance is realized only after subsequent measurements. Capturing information such as a cell line, associated genetic engineering, and time of measurements can enable downstream AI-based detection of systematic measurement biases, such as batch effects (Sprang et al, 2022), and meta-analyses across experiments. Metadata can also include interpretations of data-what were the hits in an experiment and what criteria were employed in their selection?…”