A Transcriptome Fingerprinting Assay for Clinical Immune Monitoring

Abstract

Background: While our understanding of the role that the immune system plays in health and disease is growing at a rapid pace, available clinical tools to capture this complexity are lagging.We previously described the construction of a third-generation modular transcriptional repertoire derived from genome-wide transcriptional profiling of blood of 985 subjects across 16 diverse immunologic conditions, which comprises 382 distinct modules. Results: Here we describe the use of this modular repertoire framework for the development of a targeted transcriptome fingerprinting assay (TFA). The first step consisted in down-selection of the number of modules to 32, on the basis of similarities in changes in transcript abundance and functional interpretation. Next down-selection took place at the level of each of the 32 modules, with each one of them being represented by four transcripts in the final 128 gene panel. The assay was implemented on both the Fluidigm high throughput microfluidics PCR platform and the Nanostring platform, with the list of assays target probes being provided for both. Finally, we provide evidence of the versatility of this assay to assess numerous immune functions in vivo by demonstrating applications in the context of disease activity assessment in systemic lupus erythematosus and longitudinal immune monitoring during pregnancy.Conclusions: This work demonstrates the utility of data-driven network analysis applied to largescale transcriptional profiling to identify key markers of immune responses, which can be downscaled to a rapid, inexpensive, and highly versatile assay of global immune function applicable to diverse investigations of immunopathogenesis and biomarker discovery.

BloodGen3Module: blood transcriptional module repertoire analysis and visualization using R

“…We observed a preferential restriction of this signature in one of the reference datasets constituting the BloodGen3 module repertoire annotation framework, this time suggesting a predominant role for neutrophils in driving inflammation. Finally, we have shown the utility of fixed module repertoires in the design and implementation of targeted blood transcriptional profiling assays in two other use cases—applied specifically to Covid-19 immune profiling and generic immune monitoring of immunological changes during pregnancy ( Altman et al , 2019 ; Rinchai et al , 2020b ). For this type of applications the selection of a targeted panel is carried out across the entire repertoire to avoid oversampling of a dominant signature, as may be the case otherwise when candidates are selected from ‘bulk’ differentially expressed gene lists.…”

Design of a targeted blood transcriptional panel for monitoring immunological changes accompanying pregnancy

“…To select the 8 HK genes, (i) four genes were included as HK genes in a generic 272 gene panel that we have designed and implemented previously ( 27 ); (ii) two genes were selected based on low coefficients of variation (%CV) in the MSP and PROMISSE datasets; and (iii) the last two genes were selected based on earlier publication reports.…”

Comparisons of the immunological landscape between COVID-19, influenza, and respiratory syncytial virus patients by clustering analysis

“…We collected eight gene expression profiling datasets, including four COVID-19 RNA-seq datasets [12] , [13] , [14] , [15] , one RSV RNA-seq dataset from the Gene Expression Omnibus database (GEO, https://www.ncbi.nlm.nih.gov/geo/), one RSV microarray dataset [16] , one influenza microarray dataset from GEO, and one microarray dataset for both RSV and influenza [17] ( Table 1 ). First, we collected raw data of these datasets, composed of 1,036 patients and 273 healthy controls.…”