Table 4.
Other Omics-identified sarcoidosis biomarkers.
| Omics | Tissue/assay/country | Cohort | ML or Software | Key results | Validation |
|---|---|---|---|---|---|
| Transcriptomics (61) | Tregs of PBMC and BAL/FACS, open array, PCR/Poland | 45 PS | PCA, networks | SA-related miR-155, miR-223. PS have increased TLR/NOD signaling, intrinsic apoptosis, inflammation. Upregulated TLR2/MyD88 in different subpopulations of PBMCs | No |
| Transcriptomics (62) | BAL cells/Affymetrix and qPCR/Poland | 12 SA | Hierarchical clustering, enrichment | BAL cells have increased mRNA for ribosome biogenesis and increased lymphocytes | No |
| Transcriptomics (63) | Cardiac tissue/snRNAseq/USA, | 4 CS, 4 Ctl | Seurat, SCRNIC, GENIE3, AUCell, UMAP, RunPCA, RcisTarget, CLusterProfiler | GPNMB (transmembrane glycoprotein NMB) as a novel marker of multinucleated giant cells, additional macrophage populations. mTOR (mammalian target of rapamycin) pathway activation in HLA-DR+ and SYLT3+ macrophages is associated with proliferation | No |
| Proteomics (64) | BAL cells/2DE/Sweden | 7 SA, 7 Ctl | PCA for clustering, OPLS | 25 proteins for Fc-mediated phagocytosis and clathrin-mediated endocytosis pathways | No |
| Proteomics (65) | BAL macrophages/LC-MS/Sweden | 8 SA, 6 Ctl | Multivariate models, OPLS-DA | 15 protein panel for separating LS and non-LS. R2 = 0.9, Q=0.85 | No |
| Proteomics (66) | BAL, serum LS/LC-MS/Sweden | 11 LS, 12 nonLR,12 Ctl | PCA, OPLS | Fcγ-regulation-associated factors for diagnosis, IgG related factors for phenotyping | Yes |
| Metabolomic (58) | Saliva/NMR/France | 24 SA, 45 Ctl | OPLS | Six metabolites were altered in SA, including methanol/D, butyrate/D, lactate/U, acetate/U, and N-butyrate/U. | No |
| Metabolomic (59) | Plasma/LC-MS/USA | 67 SA: 31 LF | Multivariate models, OPLS, PCA | Discriminating metabolites involved collagen pathway metabolites, the arginine-proline pathway. p-coumaroylagmatine and palmitoylcarnitine are markers for LF. | No |
| Metabolomic (60) | Plasma/NMR/Canada | 43 PS: 30 civilians, 13 veterans | MetaboAnalyst, MetaBox, SIMCA-P, PCA, PLS-DA, OPLS-DA | Six metabolites and 33 elements differ between Vet and Civ. | No |
| Metabolomic (67) | Serum/NMR/Poland | 40 SA | OPLS-DA | Exercise decreases fatty acids, triglycerides, and total cholesterol. Lipid profile as a prognosticator for lung function recovery | No. |
| Microbiome (68) | BAL/16S rRNAseq/India | 8 SA | LEfSe for microbial markers, LDA=3 | Corynebacteriales, Corynebacterium, and Neisseria.OTUID_476 for SA | No |
SA, sarcoidosis; Ctl, controls; PS, pulmonary sarcoidosis; nonLR, non-lung fibrosis; PCA, principal component analysis; LDA, linear discriminant analysis; OPLS-DA, orthogonal partial least squares discriminant analysis; SCRNIC, single-cell regulatory network inference and clustering; LS, Löfgrens syndrome; Spe, specificity; sen, sensitivity. One of 11 studies was validated.