Caitlin Smith has a B.A. in biology from Reed College, a Ph.D. in neuroscience from Yale University, and completed postdoctoral work at the Vollum Institute.
With data from molecular biology and proteomics – just to name two examples – pouring in, researchers sometimes struggle to organize the reams of information into a framework that is biologically meaningful, logical, and convenient to work with. Organizing data into pathways helps to streamline the information in a way that makes sense. This is much easier with pathway analysis software, which usually allows visual data representations that are more intuitive to use. So what else can pathway analysis software do for you?
What is pathway analysis software?
Pathway analysis involves looking through large sets of data to identify genes or proteins that are differentially expressed with different phenotypes. Biological significance can be gleaned by comparing data sets to knowledge from the literature. Done manually, this would require the entire careers of many scientists, but pathway analysis software is designed to accomplish this easily. And increasingly, this is becoming a necessity rather than a luxury. The huge sets of data that result from high-throughput experiments are best understood when one is able to view them as a reflection of a biological system, rather than one gene or protein. The types of data that researchers commonly study with pathway analysis software include microarray expression results, maps of single nucleotide polymorphisms (SNPs), mutation maps, siRNA knockout data, and proteomics profiles.
What can pathway analysis software do?
Pathway analysis software enables you to perform tasks that would otherwise boggle the mind. For example, you can build customized pathways from your data set, then overlay information about biomarkers, drugs, or toxicity to test interactions. Using the software’s ability to pull information from the published literature, you can find biological interactions among genes or proteins of interest. You can find pathways between two molecules. Or given a group of genes, you can identify regulators that they have in common. These are only a few examples – try some software to get a sense of its possibilities for your research.
Examples of pathway analysis software
There are different types of software available, including both free and commercial varieties. Several examples will be discussed here as illustrations of the capabilities of pathway analysis software – although no single software package is better. Ultimately the best kind is the one that does what you need it to do.
MetaCore™ from GeneGo/Thomson Reuters
MetaCore is designed to analyze data from experiments using - for example - microarrays, serial analysis of gene expression (SAGE), SNPs and comparative genomic hybridization (CGH) arrays, proteomics, metabolomics, yeast 2-hybrid experiments, and others. MetaCore software contains an integrated knowledge database for analyzing experimental data and gene lists. You can take advantage of the knowledge database by interrogating the biological context of lists of genes, transcripts, or proteins, as well as lists generated from microarray experiments, next-generation sequencing, proteomics work, nuclear magnetic resonance data, or co-immunoprecipitation pull-down data, for example.
MetaCore’s database is proprietary and manually curated, and includes human protein-protein interactions, human protein-DNA interactions, and human protein and compound interactions. In addition, the MetaCore database also includes human, mouse, and rat metabolic and signaling pathways.
Ingenuity Pathways Analysis (IPA®) from Ingenuity Systems
Like MetaCore, the web-based IPA software lets you analyze data from many sources, such as gene expression microarrays, microRNA microarrays, SNP microarrays, metabolomics, proteomics, RNA-Seq experiments, and other types of experiments that generate lists of genes or chemicals. Also similar to MetaCore, IPA is based on the Ingenuity® Knowledge Base, which gives users powerful search abilities that can help to frame data sets or specific targets or biomarkers in context, within a bigger picture of biological significance. You can search for information on genes, proteins, drugs, or chemicals, and then use this information to build models of your experimental system.
IPA offers different features for different functions. For example, its ‘Data Analysis and Interpretation’ speeds the identification of key relationships, mechanisms, functions, and pathways, from the reams of data. IPA’s ‘Core Analysis’ looks at the data set and reports back on signaling and metabolic pathways or networks that are most disrupted. This can help you understand the effects of changes in mRNA, protein, or metabolite levels on well-characterized pathways; understand the effects of a set of genes on a disease phenotype; or compare pathways or phenotypes at different times, or drug doses, or populations of patients. IPA also includes ‘Metabolomics®’, ‘Tox®’, and ‘Biomarker®’ features for specialized analyses.
PathwayArchitect® from Stratagene/Agilent Technologies
PathwayArchitect (powered by avadis™) includes 100 manually-curated pathways with a database of more than 2 million biological interactions. The interactions more relevant to your data sets are found more quickly with their Relevance Interaction Network algorithm. Starting with a list of proteins, Relevance Interaction Network analysis identifies a network of proteins and small molecules that are most statistically related to the biology of the listed proteins. This is especially helpful for a biologist wanting to find binding partners in protein complexes, networks of transcriptional regulation, or a group of regulatory proteins. The PathwayArchitect software is designed for biologists, so it may be less useful for those in related but not bio-focused areas.
GenePattern from the Broad Institute
GenePattern is a genomic analysis platform with over 230 tools for analyzing gene expression, proteomics, SNP analysis, flow cytometry, RNA-seq analysis data sets. It lets you create analysis pipelines using a web-based interface. Given a list of genes, GenePattern’s KSscore uses statistical analysis to determine the positional distribution of those genes. GenePattern’s Gene Set Enrichment Analysis can determine whether a set of genes shows statistically significant differences biologically – in other words, whether they might show different phenotypes.
Each pathway analysis software package offers powerful tools. Try a few to find the one that’s best for your experiments and data. Ultimately, using more than one kind of software to analyze your data may provide new insights or lend extra confidence to your results.
The image at the top of the page is from Ingenuity Systems.