A top-down measure of gene-to-gene coordination for analyzing cell-to-cell variability

Recent technological advances, such as single-cell RNA sequencing (scRNA-seq), allow the measurement of gene expression profiles of individual cells. These expression profiles typically exhibit substantial variations even across seemingly homogeneous populations of cells. Two main different sources contribute to this measured variability: actual differences between the biological activity of the cells and technical measurement errors. Analysis of the biological variability may provide information about the underlying gene regulation of the cells, yet distinguishing it from the technical variability is a challenge. Here, we apply a recently developed computational method for measuring the global gene coordination level (GCL) to systematically study the cell-to-cell variability in numerical models of gene regulation. We simulate ‘biological variability’ by introducing heterogeneity in the underlying regulatory dynamic of different cells, while ‘technical variability’ is represented by stochastic measurement noise. We show that the GCL decreases for cohorts of cells with increased ‘biological variability’ only when it is originated from the interactions between the genes. Moreover, we find that the GCL can evaluate and compare—for cohorts with the same cell-to-cell variability—the ratio between the introduced biological and technical variability. Finally, we show that the GCL is robust against spurious correlations that originate from a small sample size or from the compositionality of the data. The presented methodology can be useful for future analysis of high-dimensional ecological and biochemical dynamics.

Accurate cancer phenotype prediction with AKLIMATE, a stacked kernel learner integrating multimodal genomic data and pathway knowledge

Advancements in sequencing have led to the proliferation of multi-omic profiles of human cells under different conditions and perturbations. In addition, many databases have amassed information about pathways and gene “signatures”—patterns of gene expression associated with specific cellular and phenotypic contexts. An important current challenge in systems biology is to leverage such knowledge about gene coordination to maximize the predictive power and generalization of models applied to high-throughput datasets. However, few such integrative approaches exist that also provide interpretable results quantifying the importance of individual genes and pathways to model accuracy. We introduce AKLIMATE, a first kernel-based stacked learner that seamlessly incorporates multi-omics feature data with prior information in the form of pathways for either regression or classification tasks. AKLIMATE uses a novel multiple-kernel learning framework where individual kernels capture the prediction propensities recorded in random forests, each built from a specific pathway gene set that integrates all omics data for its member genes. AKLIMATE has comparable or improved performance relative to state-of-the-art methods on diverse phenotype learning tasks, including predicting microsatellite instability in endometrial and colorectal cancer, survival in breast cancer, and cell line response to gene knockdowns. We show how AKLIMATE is able to connect feature data across data platforms through their common pathways to identify examples of several known and novel contributors of cancer and synthetic lethality.

Highly Accurate Cancer Phenotype Prediction with AKLIMATE, a Stacked Kernel Learner Integrating Multimodal Genomic Data and Pathway Knowledge

Advancements in sequencing have led to the proliferation of multi-omic profiles of human cells under different conditions and perturbations. In addition, several databases have amassed information about pathways and gene “signatures” – patterns of gene expression associated with specific cellular and phenotypic contexts. An important current challenge in systems biology is to leverage such knowledge about gene coordination to maximize the predictive power and generalization of models applied to high-throughput datasets. However, few such integrative approaches exist that also provide interpretable results quantifying the importance of individual genes and pathways to model accuracy. We introduce AKLI-MATE, a first kernel-based stacked learner that seamlessly incorporates multi-omics feature data with prior information in the form of pathways for either regression or classification tasks. AKLIMATE uses a novel multiple-kernel learning framework where individual kernels capture the prediction propensities recorded in random forests, each built from a specific pathway gene set that integrates all omics data for its member genes. AKLIMATE outperforms state-of-the-art methods on diverse phenotype learning tasks, including predicting microsatellite instability in endometrial and colorectal cancer, survival in breast cancer, and cell line response to gene knockdowns. We show how AKLIMATE is able to connect feature data across data platforms through their common pathways to identify examples of several known and novel contributors of cancer and synthetic lethality.

Cell Cycle Gene Sets Coordination In Multiple Myeloma and Plasma Cell Leukemia

Background In multiple myeloma (MM), unlike normal plasma cells (PC), myeloma cells retain the self-renewing potential. Majority of medullary myeloma cells regardless over-expression of cyclins D stay in the G1 phase due to pro-apoptotic and cell cycle regulatory capacity of p53 depended axis. Nevertheless, after leukemic transformation in secondary plasma cell leukemia (PCL) or de novo in case of primary PCL, bone marrow myeloma cells become highly proliferative and even presenting as circulating plasma cells in the peripheral blood. We anticipate that complex “re-setting” of cell cycle gene coordination during leukemic transformation creates required background to restore proliferation activity and breakthrough mitotic restriction points. Aims The objective of our study was to define and describe complex “re-setting” of cell cycle gene coordination in MM and PCL. Materials and Methods In total, 7 healthy donors, 6 multiple myeloma and 7 plasma cell leukemia samples enrolled in this study. The mRNA from CD138+ cells was isolated using uMACS mRNA Isolation Kit Small scale (Miltenyi Biotech) and directly amplified and labeled using WT-Ovation™ Pico RNA Amplification System Version 1.0 plus WT-Ovation™ Exon Module Version 1.0 (NuGEN). Generated SPIA-cDNA was fragmented & labeled using Encore™ Biotin Module (NuGEN). cDNA was hybridized to Affymetrix GeneChip Human Exon 1.0 ST Arrays (Affymetrix, Santa Clara, USA). All samples were labeled and scanned in a randomized order to avoid batch effects. Gen sets, connected with cell cycle regulation (GO:0045786; GO:0045787) with all direct descendants (child terms) and regulation of apoptotic process (GO:0043065; GO:0043066), were taken for the Gene Set Enrichment Analysis (GSEA) and Gene Set Differential Coordination Analysis (GSDCA). Results Comparing of PCL, MM and healthy donors revealed coordinating changes between regulation of mitosis (GO:0045839; GO:0045840), apoptosis (GO:0043065; GO:0043066) and cell cycle arrest (GO:0007050). These changes were relevant for both positive and negative regulation sets. Gene expression profiling of MM samples revealed affected early phases of cell cycle (G1 phase and G1/S transition). In PCL samples co-expression changes was associated with late phases of cell cycle (G2/M transition, S and M phase) together with severe alteration in early phases. The mechanisms controlling differential cell cycle coordination were based on bioinformatic analysis suggested to include alternative transcription start sites, exon skipping and shortening of 3'UTR. The probe sets covering the 3'UTR of CCND2 were for example significantly down regulated in plasma cells of MM and PCL as compared to healthy donors supporting the existence of a phenomenon observed in breast cancer where shortening of 3'UTR mRNA CCND2 confers higher mRNA stability leading to higher protein expression and more cells to enter the S phase. Conclusion Considering revealed coordination changes allow us to offer following statements. Expression of cell cycle positive regulators is in dynamic equilibrium with cell cycle negative regulators. We suppose that this equilibrium serves as a compensatory mechanism to oncogenic events. Despite compensation mechanisms activation, whole regulatory complex seems to be imbalanced by growing “oncogenic stress” during MM to PCL progression. This study was supported by grants NT11154, NT12130, NT13190 and the EU 6th FP to MSCNET (LSHC-CT-2006-037602), the Danish Cancer Society, the Danish Research Agency (#2101-07-0007) and the KE Jensen Foundation (2006-2008). Disclosures: No relevant conflicts of interest to declare.