Children's rare disease cohorts: an integrative research and clinical genomics initiative

While genomic data is frequently collected under distinct research protocols and disparate clinical and research regimes, there is a benefit in streamlining sequencing strategies to create harmonized databases, particularly in the area of pediatric rare disease. Research hospitals seeking to implement unified genomics workflows for research and clinical practice face numerous challenges, as they need to address the unique requirements and goals of the distinct environments and many stakeholders, including clinicians, researchers and sequencing providers. Here, we present outcomes of the first phase of the Children’s Rare Disease Cohorts initiative (CRDC) that was completed at Boston Children’s Hospital (BCH). We have developed a broadly sharable database of 2441 exomes from 15 pediatric rare disease cohorts, with major contributions from early onset epilepsy and early onset inflammatory bowel disease. All sequencing data is integrated and combined with phenotypic and research data in a genomics learning system (GLS). Phenotypes were both manually annotated and pulled automatically from patient medical records. Deployment of a genomically-ordered relational database allowed us to provide a modular and robust platform for centralized storage and analysis of research and clinical data, currently totaling 8516 exomes and 112 genomes. The GLS integrates analytical systems, including machine learning algorithms for automated variant classification and prioritization, as well as phenotype extraction via natural language processing (NLP) of clinical notes. This GLS is extensible to additional analytic systems and growing research and clinical collections of genomic and other types of data.

A deep auto-encoder model for gene expression prediction

BACKGROUND

Gene expression is a key intermediate level that genotypes lead to a particular trait. Gene expression is affected by various factors including genotypes of genetic variants. With an aim of delineating the genetic impact on gene expression, we build a deep auto-encoder model to assess how good genetic variants will contribute to gene expression changes. This new deep learning model is a regression-based predictive model based on the MultiLayer Perceptron and Stacked Denoising Auto-encoder (MLP-SAE). The model is trained using a stacked denoising auto-encoder for feature selection and a multilayer perceptron framework for backpropagation. We further improve the model by introducing dropout to prevent overfitting and improve performance.

RESULTS

To demonstrate the usage of this model, we apply MLP-SAE to a real genomic datasets with genotypes and gene expression profiles measured in yeast. Our results show that the MLP-SAE model with dropout outperforms other models including Lasso, Random Forests and the MLP-SAE model without dropout. Using the MLP-SAE model with dropout, we show that gene expression quantifications predicted by the model solely based on genotypes, align well with true gene expression patterns.

CONCLUSION

We provide a deep auto-encoder model for predicting gene expression from SNP genotypes. This study demonstrates that deep learning is appropriate for tackling another genomic problem, i.e., building predictive models to understand genotypes' contribution to gene expression. With the emerging availability of richer genomic data, we anticipate that deep learning models play a bigger role in modeling and interpreting genomics.

Epistasis analysis of microRNAs on pathological stages in colon cancer based on an Empirical Bayesian Elastic Net method

Background

Colon cancer is a leading cause of worldwide cancer death. It has become clear that microRNAs (miRNAs) play a role in the progress of colon cancer and understanding the effect of miRNAs on tumorigenesis could lead to better prognosis and improved treatment. However, most studies have focused on studying differentially expressed miRNAs between tumor and non-tumor samples or between stages in tumor tissue. Limited work has conducted to study the interactions or epistasis between miRNAs and how the epistasis brings about effect on tumor progression. In this study, we investigate the main and pair-wise epistatic effects of miRNAs on the pathological stages of colon cancer using datasets from The Cancer Genome Atlas.

Results

We develop a workflow composed of multiple steps for feature selection based on the Empirical Bayesian Elastic Net (EBEN) method. First, we identify the main effects using a model with only main effect on the phenotype. Second, a corrected phenotype is calculated by removing the significant main effect from the original phenotype. Third, we select features with epistatic effect on the corrected phenotype. Finally, we run the full model with main and epistatic effects on the previously selected main and epistatic features. Using the multi-step workflow, we identify a set of miRNAs with main and epistatic effect on the pathological stages of colon cancer. Many of miRNAs with main effect on colon cancer have been previously reported to be associated with colon cancer, and the majority of the epistatic miRNAs share common target genes that could explain their epistasis effect on the pathological stages of colon cancer. We also find many of the target genes of detected miRNAs are associated with colon cancer. Go Ontology Enrichment Analysis of the experimentally validates targets of main and epistatic miRNAs, shows that these target genes are enriched for biological processes associated with cancer progression.

Conclusion

Our results provide a set of candidate miRNAs associated with colon cancer progression that could have potential translational and therapeutic utility. Our analysis workflow offers a new opportunity to efficiently explore epistatic interactions among genetic and epigenetic factors that could be associated with human diseases. Furthermore, our workflow is flexible and can be applied to analyze the main and epistatic effect of various genetic and epigenetic factors on a wide range of phenotypes.

Electronic supplementary material

The online version of this article (10.1186/s12864-017-4130-7) contains supplementary material, which is available to authorized users.

An integrated map of structural variation in 2,504 human genomes

Structural variants are implicated in numerous diseases and make up the majority of varying nucleotides among human genomes. Here we describe an integrated set of eight structural variant classes comprising both balanced and unbalanced variants, which we constructed using short-read DNA sequencing data and statistically phased onto haplotype blocks in 26 human populations. Analysing this set, we identify numerous gene-intersecting structural variants exhibiting population stratification and describe naturally occurring homozygous gene knockouts that suggest the dispensability of a variety of human genes. We demonstrate that structural variants are enriched on haplotypes identified by genome-wide association studies and exhibit enrichment for expression quantitative trait loci. Additionally, we uncover appreciable levels of structural variant complexity at different scales, including genic loci subject to clusters of repeated rearrangement and complex structural variants with multiple breakpoints likely to have formed through individual mutational events. Our catalogue will enhance future studies into structural variant demography, functional impact and disease association.

An integrated network of microRNA and gene expression in ovarian cancer

Background

Ovarian cancer is a deadly female reproductive cancer. Understanding the biological mechanisms underlying ovarian cancer could help lead to quicker and more accurate diagnosis and more effective treatments. Both changes in microRNA(miRNA) expression and miRNA/mRNA dysregulation have been associated with ovarian cancer. With the availability of whole-genome miRNA and mRNA sequencing we now have new potentials to study these associations. In this study, we performed a comprehensive analysis of miRNA and mRNA expression in ovarian cancer using an integrative network approach combined with association analysis.

Results

We developed an integrative approach to construct a network that illustrates the complex interplay among miRNA and gene expression from a systems perspective. Our method is composed of expanding networks from eQTL associations, building network associations in eQTL analysis, and then combine the networks into an integrated network. This integrated network takes account of miRNA expression quantitative trait loci (eQTL) associations, miRNAs and their targets, protein-protein interactions, co-expressions among miRNAs and genes respectively. Applied to the ovarian cancer data set from The Cancer Genome Atlas (TCGA), we created an integrated network with 167 nodes containing 108 miRNA-target interactions and 145 from protein-protein interactions, starting from 44 initial eQTLs. This integrated network encompassed 26 genes and 14 miRNAs associated with cancer. In particular, 11 genes and 12 miRNAs in the integrated network are associated with ovarian cancer.

Conclusion

We demonstrated an integrated network approach that integrates multiple data sources at a systems level. We applied this approach to the TCGA ovarian cancer dataset, and constructed a network that provided a more inclusive view of miRNA and gene expression in ovarian cancer. This network included four separate types of interactions among miRNAs and genes. Simply analyzing each interaction component in isolation, such as the eQTL associations, the miRNA-target interactions or the protein-protein interactions, would create a much more limited network than the integrated one.