A Multidimensional Bioinformatic Platform for the Study of Human Response to Surgery

Abstract

Objective: To design and establish a prospective biospecimen repository that integrates multi-omics assays with clinical data to study mechanisms of controlled injury and healing.

Background: Elective surgery is an opportunity to understand both the systemic and focal responses accompanying controlled and well-characterized injury to the human body. The overarching goal of this ongoing project is to define stereotypical responses to surgical injury, with the translational purpose of identifying targetable pathways involved in healing and resilience, and variations indicative of aberrant peri-operative outcomes.

Methods: Clinical data from the electronic medical record combined with large-scale biological data sets derived from blood, urine, fecal matter, and tissue samples are collected prospectively through the peri-operative period on patients undergoing 14 surgeries chosen to represent a range of injury locations and intensities. Specimens are subjected to genomic, transcriptomic, proteomic, and metabolomic assays to describe their genetic, metabolic, immunologic, and microbiome profiles, providing a multidimensional landscape of the human response to injury.

Results: The highly multiplexed data generated includes changes in over 28,000 mRNA transcripts, 100 plasma metabolites, 200 urine metabolites, and 400 proteins over the longitudinal course of surgery and recovery. In our initial pilot dataset, we demonstrate the feasibility of collecting high quality multi-omic data at pre- and postoperative time points and are already seeing evidence of physiologic perturbation between timepoints.

Conclusions: This repository allows for longitudinal, state-of-the-art geno-mic, transcriptomic, proteomic, metabolomic, immunologic, and clinical data collection and provides a rich and stable infrastructure on which to fuel further biomedical discovery.

FIGURE 1.

Surgical procedures and specimen collection by primary site. Depicted is the number of surgeries by major primary site as of December 17, 2021. A total of 393 surgeries have been included in the project at this time (A). Over 41,000 specimens have been collected during the peri-operative period for this project. (B) The most common specimens collected by primary surgical site. The bar graph further illustrates the necessary flexibility in the collection protocol, according to procedure type.

FIGURE 2.

Barcoding and sample tracking. Biospecimens including but not limited to blood, urine, and tissue are collected in both clinic and the operating room by clinical research coordinators. A “parent” barcode is applied and electronically scanned into the electronic data base. If applicable, the biospecimen collection core processes each parent sample into “child” sample (s), applies unique child barcodes, enters processing information into laboratory information management system, and stores samples into the patient repository. All downstream molecular assays are managed by unique identifiers.

FIGURE 3.

QC of Population Gating. Individual cell populations such as PD-1 positive (highlighted in magenta), and CD28, CD57, CD244 positive (not highlighted) are gated manually by the flow analyst. Population fractions can be calculated as percentages of their parent populations from multiple, independent plots (1, 2, and 3). Cells stained for a particular marker are classified as marker positive and are summed (eg, 1a + 1b = PD-1positive). Ideally, 1a + 1b = 2a + 2b = 3a + 3b, but inexact manual gating can result in inconsistencies between the equivalent calculated population results. In this example: 1a + 1b 1/4 33.57, 2a + 2b = 33.64, 3a + 3b = 33.65. The QC metric calculated for each population within a given sample is simply the range between the highest calculated value and the lowest calculated value. In this experiment, the cutoff threshold is arbitrarily set to 2.0% points for any given cell population within any given sample (eg, |33.57 − 33.65| = 0.08). Applying this strategy to the entire dataset, it is trivial to automate a calculation matrix for all samples and all subpopulations to highlight outliers for re-gating analysis.

FIGURE 4.

Informatics workflow. Patient clinical data is entered into the electronic data capture system and patient sample information is entered into laboratory information management system. The quality assurance/quality control (QA/QC) tool compares this data entry and reconciles all discrepant data. Sample location can be easily queried through the barcode system for efficient sample retrieval. The query tool allows investigators to search for specific clinical, biospecimen, and assay information to construct their own study-specific cohort. Assay results are returned to the data repository and reside in the central data repository for secondary analysis. Future capabilities will include the ability for external investigators to submit sample and data queries to the central data repository to support external collaborations.

FIGURE 5.

Use case illustration of multi-omics data integration in the project. (A) Correlates transcriptomics, cytokine profiling, metabolomics, and proteomics data across the 9 preliminary patients. The x axis contains patients at pre- and postoperative time points while the y axis contains the multi-omic assay datapoints. (B) Depicts the preoperative and postoperative changes in mRNA derived from buffy coat within the 9 use case patients. Preoperative mRNA values on the right are compared to postoperative mRNA values on the left. The mRNA was sequenced using Illumina NovaSeq 6000 sequencing platform. Flow cytometry data detailing systemic (blood) immune composition changes between presurgery and postsurgery (recovery) time points (C). Relative frequencies of parent cell are displayed on y axis.

FIGURE 6.

1KPP Microbiome analysis. These results are based on either fecal (A–C) or oral (D–F) samples at the preoperative initial encounter versus the postoperative day 5–7 time points. Alpha diversity index was used as a measure of microbiome diversity based on the number of amplicon sequence variants (ASVs). In fecal samples, no change was seen at preoperative versus postoperative timepoints (A). In oral samples, a significant reduction in diversity was found in the number of ASVs after surgery (D). A heatmap of log2 ASV counts for top 10 most differentially abundant classes was derived for both fecal and oral biome (B, E) and log2-fold change in ASV counts shown for all significantly different taxa at an un-adjusted P value of 0.05 (C, F). At the individual taxonomic levels, fecal samples showed significant differential abundance at the Family level for microbiota derived from the Bacteroidea and Clostridia classes (B, C). In oral samples, more taxa were found to be differentially abundant, with most differentially abundant taxa in the Fusobacteria, Actinobacteria, and Gammaproteobacteria classes (E, F).