Systems biology or pathway-based data analysis approaches allow the identification of networks of biological entities that may collectively define mechanisms and phenotypes, especially as they relate to disease. Herein, we applied an integrative systems biology workflow to hypothesize clinically relevant biomarkers and drugs targets for Alzheimer’s disease.1 Our workflow included several in silico approaches that integrate the prioritization of disease gene signatures, the analysis of disease-gene pathways and networks, and the ranking of putative drug targets based on their novelty scores (i.e., evaluating complete novelty, condition novelty or evidence of early development). We foresee this workflow as a universal tool for the prioritization of drug targets and biomarkers in complex diseases including, cancer, diabetes and many neurodiseases.
In this session, we will be using MetaCore to compare and analyze the differentially expressed genes (DEGs) in 6 brain regions of Alzheimer’s disease patients calculated from the gene expression profiles reported in Gene Expression Omnibus (GEO) dataset GSE5281. Next, we will apply Causal Reasoning in MetaCore Key Pathway Advisor (MetaCore KPA) to identify upstream regulatory hubs that could be prioritized as drug targets and/or biomarkers. The gene and protein hits identified from the upstream key hub predictions and downstream enrichment analyses will be integrated and analyzed using the network building tools available in MetaCore to understand the underlying mechanisms. Finally, the prioritized hypotheses will be evaluated and putative drug targets will be ranked based on their novelty scores, using the Drug Research Advisor-Target Druggability (DRA-TD).2 At the end of this session we will be able to answer the following key questions:
• What pathways and process networks are potentially disrupted in Alzheimer’s disease?
• What upstream key regulatory hubs are potentially activated or inhibited in Alzheimer’s disease?
• How to integrate results from upstream and downstream analyses to generate higher confidence, clinically-relevant hypotheses about drug targets and biomarkers?
• How to evaluate the resulting hypotheses and score putative drug targets?
1) Hajjo, R & Willis, C. Systems biology approaches to omics data analysis in complex diseases. 253rd Am Chem Soc (ACS) Natl Meet (April 2-6, San Francisco) 2017, Abst BIOT 461.
2) Drug research Advisor, https://projectne.thomsonreuters.com/dra/, 2017.
DID YOU KNOW?
• Use of biomarkers in clinical studies rose 15% between 2010 and 2015, increasing from 43% of trials measuring biomarkers to 58%.
• The Integrity database from Clarivate Analytics includes a unique biomarker knowledge store that connects experimental research, drug R&D and clinical studies for pivotal insights and decision making.
• The Biomarkers Module allows you to verify quickly if any potential biomarkers you identify in MetaCore using Omics data analysis are already known in the literature, and if so, learn how and where they have been used.
In this webinar, we will discuss using the Biomarkers Module of Integrity to retrieve a list of prognostic markers of prostate cancer that have translated successfully to the clinic. We will upload the list of biomarkers into MetaCore for further analysis and associate the results with public data for genes that are differentially expressed in recurrent prostate cancer compared to non-recurrent (available in the GEO series entry GSE25136), to answer these questions:
• Which prostate cancer-specific pathways are statistically enriched with the curated prognostic markers?
• Does analysis of the public gene expression data highlight any potential new biomarkers of prostate cancer prognosis?
• What curated evidence is available to support our hypothesis?
Cancer immunotherapies importance as an integral standard of care across oncological indications continues to grow. Antibody inhibition of CTLA-4 and PD-1 enhances the antitumor immune response (1), yielding high rates of objective clinical responses and ultimately melanoma and lung cancer FDA approvals. A rising challenge for these therapies is the resistance to treatment in a subset of patients due to acquired or intrinsic mechanisms (2). Beyond mutations in the tumor cells themselves, the tumor microenvironment can play an important role in the response to these treatments. It was recently shown that when treating melanoma patients with Ipilimumab, myeloid derived suppressor cells (MDSC) infiltrate into tumor cells of resistant patients and could be a predictive biomarker for resistance(3).
We will be using MetaCore and the Data Annotation & Processing tool to calculate the differentially expressed genes in a publicly available microarray dataset and upload the results into MetaCore for analysis. The data used in this session was reported in the Gene Expression Omnibus (GEO) dataset GSE41620 which studied MDSCs taken from naïve mouse blood and from mice injected with the lymphoma RMA-S cell line. Samples were drawn from blood and tumors of the xenograph and naïve mice. Using this data Pathway Map Creator in MetaCore to answer these questions:
• How to calculate differentially expressed genes from a GEO dataset and upload this data into Metacore?
• What pathway maps are potentially disrupted by the differentially expressed genes?
• What transcription factors could be regulating a significant number of the genes?
STING, also known as the Stimulator of interferon genes, has been gaining attention lately for its potential as a target for autoimmune, cancer, and anti-viral treatments. In March of 2015 it was announced that Novartis and Aduro Biotech would codevelop a STING modulating compound and by May 2016 the first patients were being dosed in efficacy clinical trials1. Recent research has highlighted the importance of STING’s role in regulating type I interferon pathways, as well as participating in inflammatory responses. Combining the knowledge from this growing body of published research into the context of an interaction pathway can be a powerful tool for analyzing data and sharing knowledge. In these webinar, we will be using a combination of network building and the Pathway Map Creator in MetaCore to answer these questions:
• How are neighboring biological interactors around STING related to each other?
• How do I build a custom pathway map detailing the interaction network I’ve built? • How do I overlay datasets or run enrichment analysis with these novel pathway maps?
Quite often in drug development, a therapy will face a number of challenges including lack of efficacy, resistance, or adverse events that restrict its broader usage in treating patients. Pathway analysis of omics data can address these challenges by reconstructing the molecular mechanisms of the therapeutic signaling cascades. Glucocorticoids are a useful class of steroids that bind to the glucocorticoid receptor and have a number of useful effects such as suppressing immune and inflammatory responses. They are currently being used to suppress an overactive immune system or inhibit excessive proliferation of lymphocytes in blood cancers. Gamma-secretase inhibitors (GSI) are effective for inhibiting NOTCH signaling, which is often overexpressed in a number of cancers and plays a major role in cell survival, differentiation, and tumorigenesis. Both of these therapies are being used or tested against T-cell acute lymphoblastic leukemia and face a number of challenges including relapse with resistance for glucocorticoids and lack of efficacy as well as toxicity for GSI. Could combining these two therapies help mitigate their individual weaknesses and improve efficacy?
I will be using MetaCore to perform pathway analysis using transcriptomic data to better understand the mechanisms behind how this combination therapy enhances the efficacy of these treatments. For this training session, we will be using the results reported in the Gene Expression Omnibus (GEO) dataset GSE33562. In this dataset, the CUTLL1 cell line was treated with PF-03084014, dexamethasone, the combination of PF-0308014, or the vehicle control to see how the cells responded to the treatment. Using this data to compare how the mono treatments compare against the combination treatment, we will seek to answer these questions:
• How is the expression signature for the combination therapy unique to the two mono therapies?
• What transcription factors are overconnected with differentially expressed genes from the combination therapy?
• How could gamma-secretase inhibitors impact glucocorticoid treatments?