Integrative multi-omics analysis reveals novel idiopathic pulmonary fibrosis endotypes associated with disease progression.

BACKGROUND: Idiopathic pulmonary fibrosis (IPF) is characterized by the accumulation of extracellular matrix in the pulmonary interstitium and progressive functional decline. We hypothesized that integration of multi-omics data would identify clinically meaningful molecular endotypes of IPF. METHODS: The IPF-PRO Registry is a prospective registry of patients with IPF. Proteomic and transcriptomic (including total RNA [toRNA] and microRNA [miRNA]) analyses were performed using blood collected at enrollment. Molecular data were integrated using Similarity Network Fusion, followed by unsupervised spectral clustering to identify molecular subtypes. Cox proportional hazards models tested the relationship between these subtypes and progression-free and transplant-free survival. The molecular subtypes were compared to risk groups based on a previously described 52-gene (toRNA expression) signature. Biological characteristics of the molecular subtypes were evaluated via linear regression differential expression and canonical pathways (Ingenuity Pathway Analysis [IPA]) over-representation analyses. RESULTS: Among 232 subjects, two molecular subtypes were identified. Subtype 1 (n = 105, 45.3%) and Subtype 2 (n = 127, 54.7%) had similar distributions of age (70.1 +/- 8.1 vs. 69.3 +/- 7.6 years; p = 0.31) and sex (79.1% vs. 70.1% males, p = 0.16). Subtype 1 had more severe disease based on composite physiologic index (CPI) (55.8 vs. 51.2; p = 0.002). After adjusting for CPI and antifibrotic treatment at enrollment, subtype 1 experienced shorter progression-free survival (HR 1.79, 95% CI 1.28,2.56; p = 0.0008) and similar transplant-free survival (HR 1.30, 95% CI 0.87,1.96; p = 0.20) as subtype 2. There was little agreement in the distribution of subjects to the molecular subtypes and the risk groups based on 52-gene signature (kappa = 0.04, 95% CI= -0.08, 0.17), and the 52-gene signature risk groups were associated with differences in transplant-free but not progression-free survival. Based on heatmaps and differential expression analyses, proteins and miRNAs (but not toRNA) contributed to classification of subjects to the molecular subtypes. The IPA showed enrichment in pulmonary fibrosis-relevant pathways, including mTOR, VEGF, PDGF, and B-cell receptor signaling. CONCLUSIONS: Integration of transcriptomic and proteomic data from blood enabled identification of clinically meaningful molecular endotypes of IPF. If validated, these endotypes could facilitate identification of individuals likely to experience disease progression and enrichment of clinical trials. TRIAL REGISTRATION: NCT01915511.

DiSNEP: a Disease-Specific gene Network Enhancement to improve Prioritizing candidate disease genes.

Biological network-based strategies are useful in prioritizing genes associated with diseases. Several comprehensive human gene networks such as STRING, GIANT and HumanNet were developed and used in network-assisted algorithms to identify disease-associated genes. However, none of these networks are disease-specific and may not accurately reflect gene interactions for a specific disease. Aiming to improve disease gene prioritization using networks, we propose a Disease-Specific Network Enhancement Prioritization (DiSNEP) framework. DiSNEP first enhances a comprehensive gene network specifically for a disease through a diffusion process on a gene-gene similarity matrix derived from disease omics data. The enhanced disease-specific gene network thus better reflects true gene interactions for the disease and may improve prioritizing disease-associated genes subsequently. In simulations, DiSNEP that uses an enhanced disease-specific network prioritizes more true signal genes than comparison methods using a general gene network or without prioritization. Applications to prioritize cancer-associated gene expression and DNA methylation signal genes for five cancer types from The Cancer Genome Atlas (TCGA) project suggest that more prioritized candidate genes by DiSNEP are cancer-related according to the DisGeNET database than those prioritized by the comparison methods, consistently across all five cancer types considered, and for both gene expression and DNA methylation signal genes.

mirPLS: a partial linear structure identifier method for cancer subtyping using microRNAs.

MOTIVATION: MicroRNAs (miRNAs) are small non-coding RNAs that have been successfully identified to be differentially expressed in various cancers. However, some miRNAs were reported to be up-regulated in one subtype of a cancer but down-regulated in another, making overall associations between these miRNAs and the heterogeneous cancer non-linear. These non-linearly associated miRNAs, if identified, are thus informative for cancer subtyping. RESULTS: Here, we propose mirPLS, a Partial Linear Structure identifier for miRNA data that simultaneously identifies miRNAs of linear or non-linear associations with cancer status when non-linearly associated miRNAs can then be used for subsequent cancer subtyping. Simulation studies showed that mirPLS can identify both non-linearly and linearly outcome-associated miRNAs more accurately than the comparison methods. Using the identified non-linearly associated miRNAs much improves the cancer subtyping accuracy. Applications to miRNA data of three different cancer types suggest that the cancer subtypes defined by the non-linearly associated miRNAs identified by mirPLS are consistently more predictive of patient survival and more biological meaningful. AVAILABILITY AND IMPLEMENTATION: The R package mirPLS is available for downloading from https://github.com/pfruan/mirPLS. SUPPLEMENTARY INFORMATION: Supplementary data are available at Bioinformatics online.

Detection of Epigenetic Field Defects Using a Weighted Epigenetic Distance-Based Method.

Identifying epigenetic field defects, notably early DNA methylation alterations, is important for early cancer detection. Research has suggested these early methylation alterations are infrequent across samples and identifiable as outlier samples. Here we developed a weighted epigenetic distance-based method characterizing (dis)similarity in methylation measures at multiple CpGs in a gene or a genetic region between pairwise samples, with weights to up-weight signal CpGs and down-weight noise CpGs. Using distance-based approaches, weak signals that might be filtered out in a CpG site-level analysis could be accumulated and therefore boost the overall study power. In constructing epigenetic distances, we considered both differential methylation (DM) and differential variability (DV) signals. We demonstrated the superior performance of the proposed weighted epigenetic distance-based method over nonweighted versions and site-level EWAS (epigenome-wide association studies) methods in simulation studies. Application to breast cancer methylation data from Gene Expression Omnibus (GEO) comparing normal-adjacent tissue to tumor of breast cancer patients and normal tissue of independent age-matched cancer-free women identified novel epigenetic field defects that were missed by EWAS methods, when majority were previously reported to be associated with breast cancer and were confirmed the progression to breast cancer. We further replicated some of the identified epigenetic field defects.

Bi-level feature selection in high dimensional AFT models with applications to a genomic study.

We propose a new bi-level feature selection method for high dimensional accelerated failure time models by formulating the models to a single index model. The method yields sparse solutions at both the group and individual feature levels along with an expedient algorithm, which is computationally efficient and easily implemented. We analyze a genomic dataset for an illustration, and present a simulation study to show the finite sample performance of the proposed method.

Using association signal annotations to boost similarity network fusion.

MOTIVATION: Recent technology developments have made it possible to generate various kinds of omics data, which provides opportunities to better solve problems such as disease subtyping or disease mapping using more comprehensive omics data jointly. Among many developed data-integration methods, the similarity network fusion (SNF) method has shown a great potential to identify new disease subtypes through separating similar subjects using multi-omics data. SNF effectively fuses similarity networks with pairwise patient similarity measures from different types of omics data into one fused network using both shared and complementary information across multiple types of omics data. RESULTS: In this article, we proposed an association-signal-annotation boosted similarity network fusion (ab-SNF) method, adding feature-level association signal annotations as weights aiming to up-weight signal features and down-weight noise features when constructing subject similarity networks to boost the performance in disease subtyping. In various simulation studies, the proposed ab-SNF outperforms the original SNF approach without weights. Most importantly, the improvement in the subtyping performance due to association-signal-annotation weights is amplified in the integration process. Applications to somatic mutation data, DNA methylation data and gene expression data of three cancer types from The Cancer Genome Atlas project suggest that the proposed ab-SNF method consistently identifies new subtypes in each cancer that more accurately predict patient survival and are more biologically meaningful. AVAILABILITY AND IMPLEMENTATION: The R package abSNF is freely available for downloading from https://github.com/pfruan/abSNF. SUPPLEMENTARY INFORMATION: Supplementary data are available at Bioinformatics online.

Detection of epigenetic field defects using a weighted epigenetic distance-based method.

Identifying epigenetic field defects, notably early DNA methylation alterations, is important for early cancer detection. Research has suggested these early methylation alterations are infrequent across samples and identifiable as outlier samples. Here we developed a weighted epigenetic distance-based method characterizing (dis)similarity in methylation measures at multiple CpGs in a gene or a genetic region between pairwise samples, with weights to up-weight signal CpGs and down-weight noise CpGs. Using distance-based approaches, weak signals that might be filtered out in a CpG site-level analysis could be accumulated and therefore boost the overall study power. In constructing epigenetic distances, we considered both differential methylation (DM) and differential variability (DV) signals. We demonstrated the superior performance of the proposed weighted epigenetic distance-based method over non-weighted versions and site-level EWAS (epigenome-wide association studies) methods in simulation studies. Application to breast cancer methylation data from Gene Expression Omnibus (GEO) comparing normal-adjacent tissue to tumor of breast cancer patients and normal tissue of independent age-matched cancer-free women identified novel epigenetic field defects that were missed by EWAS methods, when majority were previously reported to be associated with breast cancer and were confirmed the progression to breast cancer. We further replicated some of the identified epigenetic field defects.

NEpiC: a network-assisted algorithm for epigenetic studies using mean and variance combined signals.

DNA methylation plays an important role in many biological processes. Existing epigenome-wide association studies (EWAS) have successfully identified aberrantly methylated genes in many diseases and disorders with most studies focusing on analysing methylation sites one at a time. Incorporating prior biological information such as biological networks has been proven to be powerful in identifying disease-associated genes in both gene expression studies and genome-wide association studies (GWAS) but has been under studied in EWAS. Although recent studies have noticed that there are differences in methylation variation in different groups, only a few existing methods consider variance signals in DNA methylation studies. Here, we present a network-assisted algorithm, NEpiC, that combines both mean and variance signals in searching for differentially methylated sub-networks using the protein-protein interaction (PPI) network. In simulation studies, we demonstrate the power gain from using both the prior biological information and variance signals compared to using either of the two or neither information. Applications to several DNA methylation datasets from the Cancer Genome Atlas (TCGA) project and DNA methylation data on hepatocellular carcinoma (HCC) from the Columbia University Medical Center (CUMC) suggest that the proposed NEpiC algorithm identifies more cancer-related genes and generates better replication results.