Epigenetic variability in cellular identity and gene regulation
We are studying the relationship between epigenetic regulators, chromatin structure and DNA sequence and how these factors influence gene expression patterns. We recently proposed an integrative computational pipeline called HAYSTACK (https://github.com/lucapinello/Haystack). HAYSTACK is a software to study epigenetic variability, cross-cell-type plasticity of chromatin states and transcription factor motifs and provides mechanistic insights into chromatin structure, cellular identity and gene regulation. By integrating sequence information, histone modification and gene expression data measured across multiple cell-lines, it is possible to identify the most epigenetically variable regions of the genome, to find cell-type specific regulators, and to predict cell-type specific chromatin patterns that are important in normal development and differentiation or potentially involved in diseases such as cancer.
Computational methods for genome editing
Recent genome editing technologies such as CRISPR/Cas9 are revolutionizing functional genomics; however, computational methods to analyze and extract biological insights from data generated with these powerful assays are still in an early stage and without standards. We embraced this revolution by developing cutting-edge computational tools to quantify and visualize the outcome of CRISPR/Cas9 experiments. We created a novel computational tool called CRISPResso (http://github.com/lucapinello/CRISPResso), an integrated software pipeline for the analysis and visualization of CRISPR/Cas9 outcomes from deep sequencing experiments, as well as a user friendly web application that can be used by non-bioinformaticians (http://crispresso.rc.fas.harvard.edu). In collaboration with the groups of Daniel Bauer and Stuart Orkin, we recently applied CRISPResso and other computational strategies to aid the development of an in-situ saturating mutagenesis approach for dissecting enhancer functionality in the blood system with the aim of developing potential therapeutic genome editing applications for haemoglobin disorders.
Exploring single cell gene expression variation in development and cancer
Cancer often starts from mutations occurring in a single cell that results in a heterogeneous cell population. Although traditional gene expression assays have provided important insights into the transcriptional programs of cancer cells, they often measure a combined signal of a mixed population of cells and hence do not provide enough information, especially in small subpopulations of malignant cells. However, emerging single cell assays are now offering exciting opportunities to isolate and study individual cells and sub-populations in heterogeneous cancer tissues, allowing us to investigate how genes transform one subpopulation into another. Characterizing stochastic variation at the single cell level is crucial to understand how healthy cells use variation to modulate their gene expression programs and how these patterns of variation are disrupted in cancer cells. By using single cell assays such as single cell RNA-seq and multiplexed qPCR, we are developing tools to model the variability of gene expression at single cell resolution, to infer cell states by profiling their transcriptome, and to detect rare cell types and track their state transitions during development.