See how to take 13,000 possible mRNA targets down to two, using IPA

IPA’s new microRNA Target Filter helps identify and prioritize mRNA targets. In this poster, IPA was used to analyze microRNA gene expression data.  IPA identified 13,000 potential mRNA targets, and then prioritized the list using biological information and biomarker information to identify just two highly relevant microRNA targets.  IPA was able to shed light on how these biomarker candidates contribute to disease progression, and helped generate a new testable hypotheses for follow-up experiments.

Figure 1.The microRNA Target Filter associates microRNA families from uploaded microRNA data with mRNA targets either experimentally determined (TarBase) or predicted (TargetScan).   Each row in the filter shows a relationship between a microRNA  family and an mRNA target.  Thus, from the initial list of 88 microRNAs  we quickly get over 13,000 mRNA targets.

Figure 1 –  click to enlarge

Figure 2. The microRNA Target Filter enables filtering on the source and confidence of the microRNA-mRNA target relationships.  Choose whether to limit targets to those that were experimentally observed or predicted and with which level of stringency.  Limiting the microRNA-mRNA target relationships to those that are experimentally observed and predicted relationships that are of high confidence the number of targets is reduced from >13,000 to <10,000.

Figure 2 – click to enlarge

Figure 3. Use biological information from the Ingenuity Knowledge Base to further refine the targets.  Add columns of information and filter the mRNA targets using biology such as which pathways the mRNAs are in, whether the target is a kinase or a transmembrane receptor, whether it is associated with a particular disease or whether it is expressed in the cell, tissue, or species that is being studied.  Because we are trying to understand and are looking for markers of melanoma disease progression, we filtered for signaling pathways involved in cancer, cell growth and development, and integrin signaling.  This reduces the number of targets from 10,000 to 30 biologically relevant targets that can be studied further.

Figure 3 – click to enlarge

Figure 4. Use experimental mRNA expression data to prioritize targets.  Add experimentally measured mRNA expression data to find targets to microRNAs measured under the same conditions.  Apply and filter over expression pairings to prioritize microRNA-mRNA target relationships based on expression changes.  Here we added the measured mRNA expression changes from the same tissue samples and look for inverse related expression changed between microRNA and mRNA.

Figure 4 – click to enlarge

Figure 5. Use IPA to understand implications of microRNA-mRNA target interactions in melanoma.  Now that we have narrowed down the mRNA targets based on biological relevance to our experimental system, we can use IPA to better explore and understand the relationships.  Here we have selected the remaining microRNA-mRNA target pairs and added them to a pathway canvas and overlaid the mRNA expression data.  Red indicates up-regulated expression levels while green is down-regulated expression.

Figure 5 – click to enlarge

Figure 6.  Use signaling pathways for orientation.  We used the pathway tool Path Explorer to make connections between microRNAs and their targets and overlaid the microRNA expression data from metastatic melanoma.  We overlaid signaling pathways to get more familiar with the resulting pathway.  mRNAs that we measured and targets of our microRNAs are involved in melanoma and melanocyte signaling pathways.

Figure 6 – click to enlarge

Figure 7.  Identify known biomarkers for melanoma.  We overlaid biomarker information from the Ingenuity Knowledge Base onto the pathway to determine whether any of the molecules from our datasets were already identified as markers for melanoma.  We learned that several of our experimentally measured microRNAs and mRNAs are biomarkers for diagnosis, prognosis and disease progression of melanoma.

Figure 7 – click to enlarge

Figure 8.  Comparing expression levels from primary and metastatic tumor identifies two key targets.  We overlaid the mRNA expression values measured from primary and metastatic tumor and saw that microRNA targets KIT and MC1R are down-regulated in primary tumors but up-regulated in metastatic tumors.  Both are involved in the Melanocyte Development and Pigmentation Signaling Pathway and are in clinical trials as a biomarker for melanoma diagnosis.

Figure 8 – click to enlarge

Figure 9.  Use Gene View to learn more.  Gene View provides a compilation of information about genes, including information about their regulation, disease associations, expression locations, and the affects of mutations.  We used Gene View to learn more about the two targets we identified to understand their biological role and whether they could be markers of melanoma disease progression.

Figure 9 – click to enlarge

Figure 10.  Focus on melanocyte development.  Since the two microRNA targets with the highest differences in expression between primary and metastatic melanoma are associated with melanocyte development signaling pathways, we decided to look at this pathway to understand their role and how changing expression levels could lead to disease development.

Figure 10 – click to enlarge

Figure 11.  Building a testable hypothesis. The microRNA-mRNA filter identified highest priority targets based on biology and inverse correlation between microRNA and mRNA expression changes in melanoma, two important regulators in control of melanocyte signaling. Up-regulation of KIT, and MSH-R mRNA levels, may be driven by microRNAs (mir-193, mir-148, mir199A, mir218), leading to de-regulation of melanocyte development and survival.

Figure 11 – click to enlarge


  • Here we used microRNA-mRNA filter to identify and prioritize microRNA-mRNA relationships from multiple data sources that are biologically relevant to melanoma research- reduced potential targets from >13,000 to 30 to 2
  • Visualized key relationships using pathway tools to identify potential microRNA markers and the targets they may regulate
  • Gene Views gave access to biological findings and literature to understand the role of genes in melanoma disease progression
  • Built testable hypotheses for how to validate microRNAs as markers for melanoma

Return to the homepage