DBHi Bioinformatics

DBHi leads efforts in the analysis of whole genome and exome sequences. The Bioinformatics group has developed a strong infrastructure and knowledgebase, leading to many publications and grants at CHOP. Some of our collaborative work includes:

• Use of community standard and customized pipelines to drive analyses spanning raw data manipulation to variant assessment
• SNP and haplotype detection
• CNV detection and other structural variant identification
• Functional analysis of genotypes and variant regions
• Linkage to EHR-based phenotypes including analysis through Arcus

The Bioinformatics group has significant experience in the analysis of RNA-seq data from several platforms, including Illumina, PacBio, 10x Genomics, and others. The Bioinformatics Group can assist in optimal experimental design and analysis to best answer specific scientific questions using RNA-seq methods.

• Assistance in generating experimental designs for RNA-seq analyses (ANOVA, cross-over, case-control, etc.)
• Overall analysis of RNA-seq experiments starting from raw data and including quality control, read alignment, differential expression, clustering, functional genomics
• Mapping and analysis of splicing isoforms and allele-specific expression, gene fusions, and RNA editing
• Small RNA analyses (miRNA, snoRNA, circRNA, lncRNA, etc.)
• Post-alignment secondary and exploratory analyses: clustering, formal functional genomics analyses, systems and network analyses
• Single-cell RNA-seq analyses: cell clustering and trajectories, RNA velocity, cluster-defining genes, heterogeneity analyses

Optical mapping is a technique that generates “fingerprints” of long-read DNA sequence that can be used to compare strand-by-strand structure over many kilobases, thus revealing large-scale structural features such as inversions, deletions, and insertions specific to each genome tested. The Bioinformatics group can assist in design and implementation of optical mapping experiments.

• Analysis of optical mapping data generated with the Bionano platform
• Integration of optical mapping and whole-genome sequencing or other structural analyses

Proteomics is the global analysis of proteins within a biological system. Proteomics approaches use mass spectrometry to characterize and quantify protein abundances, structures, regulation, modifications, interactions, etc. The Bioinformatics group works with the CHOP Proteomics Core to design and analyze proteomics experiments.

• Experimental design and functional analysis of proteomics data (DDA, DIA, and PRM)
• Proteogenomics analysis by integrating RNA-seq and proteomics data
• Analyzing immunopeptidome data to identify MHC/ HLA peptides
• Downstream analysis of proteomics data such as differential expression, clustering, and network analysis

• Analysis of ChIP-seq data to identify protein-DNA interactions
• Analysis of Methyl-seq data to identify DNA methylation
• Chromosomal Confirmation analysis to study spatial organization of chromatin
• miRNA and other small RNA analysis

DBHi supports the analysis of expression array data from multiple manufacturers and technologies. In addition to providing outputs from appropriate multivariate statistical tests and visualizations, the Bioinformatics group has an extensive toolset to analyze markers and copy number.

• Normalization and summarization of array data
• Significance Testing identify differential expression
• Predictive modeling and classification
• Pathway and ontology enrichment

Bioinformatic analysis is often considered from the perspective of a single platform analysis such as RNA-seq, WGS, arrays, etc. However, the true power of computation is when multiple approaches are used to attack a problem from multiple angles. A simple example is complementing WGS with RNA-seq; this approach can identify a variant/mutation and consequences simultaneously.

• Expertise in parallelized study design and statistical analyses
• Expertise in platform data integration and analyses
• Experience in working with consortia and multi-group projects

After most bioinformatics analyses, it is often useful to do context-specific analyses to identify pathways and functions related to the genes and/or markers associated with the experiment. The Bioinformatics group has tools to enhance knowledge of molecular pathways and gene function to provide both a statistical and biological basis of significance, and applies those tools to identify robust biomarkers of pediatric disease.

• CHOP researcher license availability for bioinformatics packages and tools such as Ingenuity Pathway Analysis (IPA) and Metacore
• Expertise in analysis of systems biology networks
• Machine learning and other statistical approaches to classify pathways, experimental biomarkers, etc.

The Bioinformatics group provides quantification of phenotype similarity for analyzing complex disease cohorts and metabolomics modeling linking gene expression levels to metabolics errors and consequent diseases.

• FBA
• COBRA
• CORDA
• SCEDAR

The Bioinformatics group creates custom scientific websites, pipelines, software package and methods development. They also develop intra- and internet interfaces for scientific data delivery and exploration, such as through customization of ready-made interfaces (R Shiny) or through developing new websites or portals.