- What is Systems Biology?
- What are Dynamic Bayesian Networks?
- What type of data can Seralogix incorporate into a model?
- What format does the data need to be in?
- What kind of results can I expect?
- What will the web report "output" look like?
- Is there a demo of the analysis results available?
- Can Seralogix analyze data from the new next generation sequencing technologies such as by Illumina, Rosch 454, PacBio, Ion Torent?
- How does BioSignatureDS differ from bioinformatics software?
- What is the computational workflow/process?
- How long will it take?
- What prior knowledge is incorporated into the analysis?
- How robust are the models?
- How sensitive are the models?
- What type of models can be developed?
- What is a mechanistic gene? How is it defined?
- What is a Gene Ontology category?
- Wow! I am interested in your solutions, what do I do next?
What is Systems Biology?
Systems Biology is the study of a cell, tissue or organism at the system level. It provides an integrated and interacting view of genes, proteins and biochemical reactions. Instead of analyzing individual system components (i.e. genes, proteins and their functions), these biologists consider a system in its entirety including all relevant components and their interactions. Because of the complexity of biological systems, systems biology must incorporate traditional hypothesis-driven research with computational discovery-driven research.
Seralogix analysis incorporates all types of your data including gene expression data, protein data and metabolite profiling data. Your data is analyzed in the context of existing biological knowledge to identify the relevant mechanistic pathways affected and how they change over time.
Bayesian networks are graphical models that represent conditional dependencies and independencies among the variables corresponding to biological measurements. These variables are illustrated using nodes that are connected together by lines which represent the relationships between variables. Figure 1a is an example of a Bayesian network describing a gene regulation network. Each gene's expression is represented by a variable that describes how the genes are regulated by each other. This analysis can become confusing when only a few variables are being analyzed, but the graphical representation illustrates where the regulatory relationships exist between the genes.
Dynamic Bayesian networks are Bayesian networks that are capable of incorporating temporal processes such as time series and feedback loops, essential features of most biological systems, as illustrated in Figure 1b. Thus the ability to incorporate experimental time series measurements is particularly important for modelling biological networks.
What type of data can Seralogix incorporate into a model?
Seralogix can incorporate most types of biological data into a model; including, but not limited to:
- Microarray data
- Next generation rnaSEQ data
- Next generation dnaSEQ data for variant analysis
- miRNA data
- siRNA data
- Massively parallel sequencing data
- Proteomics (mass spec) data
- Protein chip data
- Physiological data
- Time course data
What format does the data need to be in?
Most common data formats are accepted, including:
- GenePix Results (gpr)
- Affymetrix (GCOS) results
- XML-derived formats including MAML
- Excel and other common formats used for protein or physiological measurement data
- Fastq, etc.
What kind of results can I expect?
The resulting data you receive depends on the type of service you have selected, and may include:
- Temporal modeling of gene and gene set expression
- Classical statistical analysis
- Comprehensive gene ontology (GO) anaylsis and clustering
- Chromosomal mapping of differentially signficant genes
- Pathway analysis with fully interactive visualization and antimation
- Identification of mechanistic genes underlying genetic relationships
- Disease models based on dynamic Bayesian models suitable for simulation and what-if analysis
- Disease models suitable for pattern recognition applications (diagnostics)
- Regulatory network structure learning from perturbation studies
- Interactome models and results for infectious diseases or other interacting drugs/toxins
What will the web report "output" look like?
The results are posted on Seralogix's secure client portal, accessible from any computer connected to the internet. On the results page, you will find tabs detailing various aspects of the analysis. Depending on the service chosen, analysis results will include for example:
- Background – data provided to Seralogix, assumptions, and other considerations
- Methods – description of the analysis
- Gene List – details for genes where expression has changed between treatments, mapped to their gene ontology
- Gene Ontology – gene ontology group activation analysis, mechanistic gene identification, and gene ontology scoring
- Pathway Analysis & Models – candidate pathways for further analysis, pathway scoring, subnet analysis, and comparative analysis
- Chromosome mapping – a figure was loading, followed by an error message; what will this tab contain/show?
- Biological System Disease model – a graphic of the disease model is presented along with options of searching for genes within the model by name, description(function), or experimental condition. Mechanistic genes can be highlighted within the model. Time-series studies can be visualized with animation
- Mechanistic Genes – all mechanistic genes are listed by timepoint and include a description and source. Clicking on a gene reveals additional information about the gene including its class, sequence similarity database (SSDB) motif, database accession numbers, position, amino acid, and nucleotide sequences
Is there a demo of the analysis results available?
Yes, there is a demo available to see here, or you can also see more details of our complete analysis process by going to our computational approach introduction and/or computational pipeline details.
Can Seralogix analyze data from the new next generation sequencing technologies such as by Illumina, Rosch 454, PacBio, Ion Torent?
Yes. Seralogix can help you understand your next generation (massively parallel) sequencing data, maximizing the return on your investment. Because next generation sequencing technologies utilize gigabase-scale throughput and short read lengths, they are problematic for conventional data analysis techniques. Seralogix's sequence pipelining service allows clients to view and analyze sequence alignments, nucleotide variants, and splice isoforms, and perform cross-sample comparisons. For more information, see What type of data can Seralogix incorporate into a model.
How does BioSignatureDS differ from bioinformatics software?
Other bioinformatics software allows only a limited view of a biological system; only Seralogix's proprietary approach offers the complete picture. Our next generation solutions enable biological discoveries that are otherwise not possible. Highly flexible and easily customized, our pipeline approach integrates varied experimental data sets including time-course data, with each other and with existing biological knowledge. Unlike other analyses, our algorithms use powerful Dynamic Bayesian methods of machine learning and pattern recognition to decode the complexities of systems biology.
What is the computational workflow/process?
See our computational pipeline for details on each step of the process.
How long will it take?
The turn around time for results is dependent on many factors, including the amount of data to be analyzed, and the condition of the data. Once the data is formatted properly for input into the software, most analyses take 10 days or less.
What prior knowledge is incorporated into the analysis?
Seralogix's BioSignature Analysis™ incorporates data from public databases including:
- Affymetrix NetAffx
- NCBI GenBank
- Gene Ontology Consortium
- Numerous other pulbic databases
How robust are the models?
The robustness of a model is dependent on the quality and quantity of the data as well as certain assumption that may be taken in preprocessing the experimental data for modeling. We can run tests that allow us to measure the robustness and discriminatory power of the models by pathway or other disease model.
How sensitive are the models?
Here again the sensitivity of a model is dependent on the quality and quantity of the data. We can tune models to be for sensitivity but selectivity may be compromised. We can run tests that allow us to measure the sensivitivity and selectivity of the models by pathway or other disease model.
What type of models can be developed?
What type of models can be developed?
Models can be developed based on:
- Sub-cellular localization
What is a mechanistic gene? How is it defined?
Mechanistic genes are genes whose products control key regulatory points in pathways through a variety of methods including altering gene transcription or translation or through post-translation processes such as protein phosphorylation. Traditional analysis only identifies the genes that have changed. However, subtle alterations of the expression of mechanistic genes can induce substantial effects on biological processes. In order to identify these mechanistic genes, all genes must be analyzed in context of their parent and neighboring genes. Dynamic Bayeisan networks provides the techniques which can analyze relationships and not just a change in state of one gene. Refer to our mechanistic model analysis and interrogation descriptionfor additional details and examples.
What is a Gene Ontology category?
Gene Ontology (GO) categories are species-independent, qualitative attributes that provide a classification of gene products into molecular functions, biological processes, and cellular components to describe attributes of gene products. Molecular function describes the biochemical function of a gene product; biological process describes a broad biological objective; and cellular component describes the location of a gene product within cellular structures and macromolecular complexes.