Multilayer systems medicine analysis of chr1q amplification in MM. (A) 2D coamplification (cyan) and 3D Hi-C contact (red) maps of chr1q locus in MM cells used to identify TCDs and TADs, respectively. Map overlay identified 4 major coamplified domains that retain a preserved 3D structure (B1-B4 hyperdomains). (B) Schematic overview of the analysis strategy used to detect candidate gene drivers of high-risk phenotypes in chr1q-amp MM. Scanning across the 2215 genes in chr1q, those fulfilling the following criteria were considered candidate drivers: (1) their genetic amplification (amplif.) is significantly associated with poor prognosis (MMRF data set, N = 896); (2) their genetic amplification is significantly associated with overexpression (MMRF data set, N = 896); (3) overexpression (overexpr.) is associated with poor prognosis in the MMRF (N = 896) and Arkansas data sets (N = 413); (4) significant epigenetic activation (ie, H3K27ac gain) is detected in chr1q-amp vs nonamplified samples (Jin2018,³⁴ n = 10). The B1-B4 hyperdomains were also used as a reference here (5).³⁰ Overall, our analysis identified 103 genes across the chr1q arm as candidate drivers for a high-risk MM prognosis. (C) Analysis overview. From top to bottom: chr1q cytogenetic map; copy number profiles of chr1q genes across MMRF patients detecting whole-arm amplification (amplif.; ∼29%), partial amplification (∼7%), no amplification (∼63%), and deletions (∼1%); survival analysis of genetic amplification of chr1q genes across MMRF patients (WGS, 73 genetic parameters; dark green bars, P values; light green bars, hazard ratio; gray bars, percentage bootstrapping confidence levels); Pearson correlation analysis between copy number ratios (WGS) and expression (RNA-seq; blue bars indicate Pearson correlation P values); survival analysis of chr1q gene expression (RNA-seq) in MMRF (brown) and Arkansas (yellow) data sets (bars indicate analysis P values); differential H3K27ac analysis between chr1q-amp (n = 5) vs nonamplified (n = 5) MM cells (red bars indicate differential log2 fold-change enrichment scores); 4 chr1q domains (B1-B4) with conserved TAD/TCD structures; candidate pathogenic driver genes (n = 103, pink bars) identified by the current analysis (the previously known MCL1, ARNT, ILF2, and CKS1B genes are shown here). (D) Analysis overview of candidate driver genes (103) across chr1q cytogenetic bands. Distribution of WGS multivariate analysis scores (-log₁₀P value; upper panel) and percentage of candidate genes (relative to band gene density; lower panel) per cytogenetic band. The highest candidate gene density was detected in 1q22 and 1q23.3 bands (highlighted here), with 1q23.3 also displaying the highest survival significance scores. (E) The PBX1 gene as a prominent candidate occupying alone a single TAD, displays strong epigenetic activation across PBX1 body and putative enhancers in chr1q-amp myeloma PCs.