Multi-omics subtyping pipeline for chronic obstructive pulmonary disease

Abstract

Chronic Obstructive Pulmonary Disease (COPD) is the third leading cause of mortality in the United States; however, COPD has heterogeneous clinical phenotypes. This is the first large scale attempt which uses transcriptomics, proteomics, and metabolomics (multi-omics) to determine whether there are molecularly defined clusters with distinct clinical phenotypes that may underlie the clinical heterogeneity. Subjects included 3,278 subjects from the COPDGene cohort with at least one of the following profiles: whole blood transcriptomes (2,650 subjects); plasma proteomes (1,013 subjects); and plasma metabolomes (1,136 subjects). 489 subjects had all three contemporaneous -omics profiles. Autoencoder embeddings were performed individually for each -omics dataset. Embeddings underwent subspace clustering using MineClus, either individually by -omics or combined, followed by recursive feature selection based on Support Vector Machines. Clusters were tested for associations with clinical variables. Optimal single -omics clustering typically resulted in two clusters. Although there was overlap for individual -omics cluster membership, each -omics cluster tended to be defined by unique molecular pathways. For example, prominent molecular features of the metabolome-based clustering included sphingomyelin, while key molecular features of the transcriptome-based clusters were related to immune and bacterial responses. We also found that when we integrated the -omics data at a later stage, we identified subtypes that varied based on age, severity of disease, in addition to diffusing capacity of the lungs for carbon monoxide, and precent on atrial fibrillation. In contrast, when we integrated the -omics data at an earlier stage by treating all data sets equally, there were no clinical differences between subtypes. Similar to clinical clustering, which has revealed multiple heterogenous clinical phenotypes, we show that transcriptomics, proteomics, and metabolomics tend to define clusters of COPD patients with different clinical characteristics. Thus, integrating these different -omics data sets affords additional insight into the molecular nature of COPD and its heterogeneity.

Fig 1. Flow chart illustrating -omics subtyping pipeline.

T, P, and M denote the Transcriptomic, Proteomic or Metabolomic data respectively. The sample size (N) and number of features are provided for each of the individual data sets, including the Clinical data. In the first step, uninformative features were filtered by only moving forward with features that had at least one association with demographic, clinical, and imaging variables (but excluding common variables such as age and sex, as well as clinical blood count variables; p = 242 clinical variables). In the second step, dimension reduction was performed using Autoencoders where EM represents embeddings (or using PCA—not shown). In the third step, the reduced dimensions were used to cluster subjects into subtypes using subspace clustering (or k-means—not shown). Finally, the subject subtypes were evaluated in the fourth step by identifying the features (transcripts, proteins, metabolites) or clinical variables (using the complete set p = 273 of clinical variables) that discriminated between the subtypes.

Fig 2. Normalized Jaccard similarity between the clusterings of different datasets.

T, P, and M, represent the Transcriptomic, Proteomic or Metabolomic results respectively. The normalized Jaccard accounts for the varying sample sizes for each subtype.

Fig 3. Flow chart for multi-omics pre- and post-clustering integration.

T, P, and M, represent the Transcriptomic, Proteomic or Metabolomic data sets respectively. Tran, Prot, and Met represent the Transcriptomic, Proteomic or Metabolomic subtypes respectively. 1 and 2 signify the large or small subtype respectively.

Fig 4. Summary of post-clustering -omics integration.

A) Upset plot representing the intersecting clusterings from the single-omic analysis among subjects with all 3 -omic profiles. B) A Venn diagram of the smaller clusters for comparison among subjects specifically categorized in the smaller cluster in only 1 of the 3 -omic clusterings.