Project description:As Asians are underrepresented across many omics databases limiting the potential of precision medicine of the global population. It is thus important for multi-omics derived quantitative trait loci (QTLs) to fill the knowledge gap of complex traits in the Asian ancestry. Integrating omics data from genomics and epigenomics (including DNA methylation and RNA-seq from blood), we developed iMOMdb, an open-acesss database to enhance disease prediction models and precision medicine outcomes in Asian pregnant women.
Project description:One challenge of cancer precision medicine is the heterogeneity of genetic and non-genetic alterations that result in dysfunctional molecular pathways. As an emerging drug discovery effort, dysregulation in Hippo pathway signaling is known to drive oncogenesis across numerous cancer types but lacks recurrent mutation(s) that are often found in other canonical signaling pathways. Here, we use first principles approach to develop a machine-learning framework to identify a robust, lineage-independent gene expression signature to quantify Hippo pathway dependency in cancers. Through integrating data from multi-omics platforms, this data-driven approach has enabled identifying a proposed combination with MAPK inhibition for direct targeting of Hippo pathway dependent cancers for which we then elucidate the underlying molecular mechanism. The results underscore how a multifaceted approach, computational models combined with laboratory efforts, can accelerate precision medicine efforts toward co-targeting Hippo and MAPK pathways, an approach that can be generalized to other key cancer signaling pathways to define therapeutic strategies.
Project description:One challenge of cancer precision medicine is the heterogeneity of genetic and non-genetic alterations that result in dysfunctional molecular pathways. As an emerging drug discovery effort, dysregulation in Hippo pathway signaling is known to drive oncogenesis across numerous cancer types but lacks recurrent mutation(s) that are often found in other canonical signaling pathways. Here, we use first principles approach to develop a machine-learning framework to identify a robust, lineage-independent gene expression signature to quantify Hippo pathway dependency in cancers. Through integrating data from multi-omics platforms, this data-driven approach has enabled identifying a proposed combination with MAPK inhibition for direct targeting of Hippo pathway dependent cancers for which we then elucidate the underlying molecular mechanism. The results underscore how a multifaceted approach, computational models combined with laboratory efforts, can accelerate precision medicine efforts toward co-targeting Hippo and MAPK pathways, an approach that can be generalized to other key cancer signaling pathways to define therapeutic strategies.
Project description:we conducted integrative multiple levels of omics data including transcriptome, phosphoproteome, proteome and metabolome in different time-course of sepsis-associated liver dysfunction (SALD). This is the first trial to suggest the statistical pathway of integrative multi-omics data in sepsis. Given the increasing number of studies collecting multi-omics data but limited overview of the methodological framework for integrative analyses (Liu, Ding et al. 2013, Petersen, Zeilinger et al. 2014, Shah, Bonder et al. 2015), integrative approach in sepsis with liver dysfunction in this study will provide novel insights into the development of sepsis and ultimately offer new tools for overcoming the present diagnostic limitation. Therefore, a combined multi-omics dataset will give better accessibility of adoption in disease, and insight to identify the promising candidates for therapeutic strategies.
Project description:We recently defined a gene expression-based signature of high-risk multiple myeloma; this predictive signature was developed with and independently validated for newly diagnosed patients treated with high dose therapy and stem cell rescue. Here we use Phase 3 clinical trial data to show that this signature also predicts short survival in relapsed disease treated with single agent bortezomib or high dose dexamethasone. In addition, a survival signature derived with relapsed myeloma samples identified newly diagnosed patients with short survival. Taken together these data suggest that a similar biology underlies poor outcome in both newly diagnosed and relapsed myeloma and provide strong evidence that the high-risk signature is a powerful tool to identify patients who are candidates for new therapeutic regimens. Keywords: Model validation See above (Series_summary)
Project description:In this work, we used an integrative, multi-omics approach and systems-level analysis to generate a mechanistic disease model for EOFAD in an hiPSC-derived neuron model system.
Project description:In this work, we used an integrative, multi-omics approach and systems-level analysis to generate a mechanistic disease model for EOFAD in an hiPSC-derived neuron model system.
Project description:In this work, we used an integrative, multi-omics approach and systems-level analysis to generate a mechanistic disease model for EOFAD in an hiPSC-derived neuron model system.
