Modeling Splicing in normal tissues and neurodegenerative disease

University Of Pennsylvania, Philadelphia PA

Investigators

Linked publications & trials

Abstract

DESCRIPTION (provided by applicant): The long-term goal of this project is to map the determinants of the human transcriptome and their effect in neurological disease. Over 90% of human genes are alternatively spliced, with tightly regulated changes in exon inclusion observed across many tissues such as brain and muscle. Splicing is associated with numerous diseases with an estimated 15 to 50 percent of human disease mutations affecting splice-site selection. Commonly, mutations occur in non-coding regions but disease studies cannot assign these function. Currently separate studies measure exon inclusion levels, binding sites of splicing regulators, and genomic variations. To fully exploit these data there is a need for methods that (a) integrate these data to identify underlying regulatory mechanisms, and (b) predict the consequences of sequence change, especially in regulatory elements in non- coding regions. To address these needs, in Phase 1 of this project we will create a human and mouse model for tissue-dependent splicing model, focusing on the central nervous system (CNS). The model will combine the abovementioned data sources to predict splicing outcome from genomic sequence and assess in silico the effect on splicing of small nucleotide variations (SNV). In Phase 2 of the project, we will collaborate with Dr. Kristen Lynch and perform elaborate biochemical experiments to validate novel regulatory mechanisms identified by the model of Phase 1, focusing on the CNS and genes involved in age-related neurological disease. Specifically, we will predict and validate disease-associated targets of two key RNA-binding proteins with CNS/disease function, TDP-43 and QKI. In Phase 3 we will collaborate with Dr. Alice Chen-Plotkin and apply Phase 1 splicing model and Phase 2 experimental validation to the study of frontaltemporal lombar degeneration (FTLD-TDP) where TDP-43 plays a key role. First, we will use the disjoint TDP-43 genomic datasets already available to produce a TDP-43 centered splicing code model that addresses key questions about TDP-43's function in disease and normal tissues. Regulatory hypotheses from the model will be tested using Phase 2 methods. Next, we will apply the TDP-43 centered code to assess the effect on splicing of genetic variations found in a cohort of 512 FTLD-TPD patients. Genetic variations predicted to effect splicing, and enriched in FTLD-TDP patients compared to a cohort of over 1000 controls will be verified using mini-gene reporter assays and/or RNA from matching patients' brain samples available in Dr. Chen-Plotkin's lab. Overall, the research proposed in this grant will create a necessary and unique framework to elucidate the determinants of FTLD-TDP and human transcriptome complexity.