Lynx Navigation Tutorial

This tutorial illustrates an example on how to traverse through Lynx and its functionalities.

From a biological perspective Autism Spectrum Disorders (ASD) are known to be associated with increased incidence of epilepsy. The example shows how a user can generate gene sets and relevant molecular mechanisms contributing to seizures in patients with Autism.

NOTE: The search results in the use case may differ from the actual results eventually due the updated databases or additional sources added.

Unable to display content. Adobe Flash is required.

Tutorial created using Jing Software on 08-20-2013.

Screenshots for Lynx navigation tutorial example

Step 1: The user can start from Search page in Lynx from the right-side corner of the front page and search for gene sets related autism. The user can perform ‘fuzzy’ search for autism candidate genes against ‘disease’ object. This results in 483 genes for autism from various sources such as OMIM, AutDB and disease database all integrated in Lynx Knowledgebase (Lynxkb).

Step 2: The search can be further refined for epileptic symptom related genes from the autism set. The genes can be further filtered using ‘seizures’ as a fuzzy search term. This results in a total of 59 genes that are shared between autism and epilepsy.

Step 3: The set of this 59 genes can then be submitted to Enrichment analysis or Network-Based Prioritization (NBP) to find relevant genes and molecular mechanisms contributing to seizures in autism.

Step 4: The enrichment page within Lynx for these set of genes showed over representation of the functional categories associated with ionotropic glutamate receptor binding, synaptic transmission and voltage-gated sodium channel all relevant categories already associated with ASD as well as epileptic phenotype.

Step 5: The user can also submit the 59 genes to Network-based Prioritization page that allows selecting high confidence genes and subnetworks relevant to user's interest.

Step 6: The query set can be either assigned a rank based on their strength of association with autism or can be assigned a flat score of '1' by default in the Network Prioritization Page. The heat-kernel ranking algorithm is used as a default algorithm with STRING 9.0 as an underlying global network for prediction. The resulting gene sets are prioritized by their P-value.

Step 7: The result set from the NBP can be again fed as an input to enrichment analysis tool. The priortized genes also show enrichment for relevant categories such as synpatic transmission, ionotropic glutamate receptor binding and voltage-gated sodium channel all molecular mechanisms again relevant to ASD with seizure phenotypes.

Step 8: The user can also download the results at every step for futher analysis.

Lynx enrichment and proritization Tutorial

This tutorial illustrates an example on how to use Lynx to perform a combo analysis on enrichment and prioritization.

We use the analysis of gene expression profiling of airway epithelial cells involved in environmental asthma to illustrate the use of existing and newly added Lynx tools. The data used in this case study is accessible at NCBI GEO database under accession GSE8190. According to the GEO metadata and corresponding article by Yang et al. (1) the airway epithelial cells were obtained via bronchial brush and bronchoalveolar lavage from 39 subjects comprising three phenotypic groups (non-atopic non-asthmatic, atopic non-asthmatic and atopic asthmatic) 4 h after instillation of lipopolysaccharide (LPS) in three distinct sub-segmental bronchi. RNA transcript levels were assessed using whole genome microarrays. To formulate a weighted hypothesis about the LPS response in airway epithelial cells, we have performed the following steps:

NOTE: The search results in the use case may differ from the actual results eventually due the updated databases or additional sources added.

Screenshots for Lynx annotation and prioritization tutorial example

Step 1:the 388 genes DE in all phenotypic conditions under investigation (control, atopy+/asthma−, atopy+/asthma+) with contrast of LPS and saline exposure were extracted from the article’s supplementary materials. By removing duplicates and correcting obsolete synonyms, we obtained a clean set of 283 genes that were used in the Lynx analysis.

Step 2: A Lynx enrichment analysis of 283 DE genes, obtained in Step 1, was performed against sixteen feature categories. The results reveal a highly significant overrepresentation of genes involved in cytokine and chemokine response pathways, such as: Cytokine Signaling in Immune system ( Reactome, 75790); Interferon Signaling ( Reactome, 25229) and NF-kappaB signaling pathway ( KEGG, hsa04064) in the set of genes under consideration. These results are consistent with the discovery presented in the source article (1) stating that the LPS stimulation resulted in pronounced transcriptional response across all subjects in airway epithelia, with strong association to nuclear factor-B and IFN-inducible genes.

Next two figures show enrichment results for 3 newly added categories, PUBMED, UNIPORT Keywords, and INTERPRO domain .

Step 3: In order to predict additional genes and subnetworks potentially involved in the inflammatory response to LPS in airway epithelia, 283 DE genes from Step 1 were analyzed by Cheetoh. This algorithm uses both features and network as an input for gene prioritization. In the aforementioned case, the enriched categories from GO and phenotype were used as features and STRING 9 was used as an underlying global network. The top ranked 100 genes , containing 23 genes from the input, were resubmitted for the enrichment analysis.

The results of the enrichment analysis against pathways databases (not used in the gene prioritization process) demonstrated a significant boost for the categories of interest. For example, Cheetoh was able to identify 21 out of 100 genes in the Chemokine category [GO:0008009] versus 9 out of 282 genes before the prioritization.

We were also able to identify the toll-like receptor signaling pathway with this prioritized gene list.

Network generated by String by submitting the top 100 genes.

(1). Yang,I.V., Tomfohr,J., Singh,J., Foss,C.M., Marshall,H.E., Que,L.G., McElvania-Tekippe,E., Florence,S., Sundy,J.S. and Schwartz,D.A. (2012) The clinical and environmental determinants of airway transcriptional profiles in allergic asthma. Am. J. Respir. Crit. Care Med., 185, 620–627.

WGCNA tutorial

This web application takes gene expression data (at least 2 samples with identifier of choice in the first column) and corresponding clinical annotations for each sample as input.

Sample expression file:

                            GeneSymbol(or other IDs),Sample1,Sample2,…,SampleN
                            A1BG,4.68,4.56,…,4.50
                            A1BG,5.28,4.23,…,5.35
                            A1CF,6.68,6.76,…,6.08
                        

Sample clinical file, entries can be binary numbers (0/1 for yes/no of the condition), or a scaler number to measure the level of a condition. Note that the clinical annotations should be transferred into numerical numbers before uploading to the server:

                            SampleID,Condition1,Condition2,…ConditionN
                            Sampl1,1,0,…,1
                            Sampl2,1,1,…,0
                            …
                            SamplN,3,1,…,3
                        

The WGCNA tool outputs the modules clustered on co-expression network as well as the correlation and Pvalue between the individual module and all available clinical annotations. Each module is assigned with an arbitrary color name. For example, in the demo example the darkorange2 module is upregulated when applying the treatment but invariant through time/culture.

All resulting gene modules can be directed to any Lynx tools for further analysis, such as enrichment, prioritization, etc. Therefore user can choose a gene module based on its correlation with one or multiple clinical condition (e.g. royalblue module in our demo is upregulated through the time change module, which results in positive correlation value in time with significant pvalue), and look at its enriched functional categories by submitting them to enrichment or get more related genes by using our prioritization tool.