Table of Contents Introduction miRmut2GO miRpair2GO

Introduction

miR2GO is an integrative web-platform for comparative analysis of microRNA function. It includes two functions: miRmut2GO and miRpair2GO. miRmut2GO is used to compare target gene sets for the reference (wild-type) and derived (mutated) alleles of miRNAs with genetic and /or somatic mutations, while miRpair2GO is used to compare target gene sets for different miRNAs.

miRmut2GO

Input

1. Select prediction method:
   Users can select one of the following miRNA target prediction methods
           TargetScan: predict miRNA targets using TargetScan.
           miRanda: predict miRNA targets using miRanda.
   Or users can combine the output of these two prediction methods by using:
           Union of the predicted target sets from TargetScan and miRanda, or
           Intersection of the predicted target sets from TargetScan and miRanda.
   
   
   

2. Specify p-value threshold for functional enrichment analysis The p-value threshold is used to determine the significantly enriched functional categories for each target gene set.

   
   
   

3. Select Gene Ontology hierarchical filtering level Hierarchical filtering is used to group the similar GO terms in the hierarchy of the GO graph. For each group of GO terms, one GO term is selected as the representative term for the group. Three hierarchical filtering options are:
   none: No hierarchical filtering, thus all the enriched GO terms are presented independently.
   moderate: Moderate filtering allows a single parent for each group in the GO hierarchy where all the enriched descendent terms of the parent term are included in the group and then the term with the lowest enrichment p-value is considered as representative term for the group.
   strong: Strong filtering allows multiple parents for the same group. For strong filtering option, the groups in the GO term hierarchies are defined by including the GO terms with common descendent in the same group. Thus the total number of representative terms are expected to be lower by using strong filtering.
   
   

4. Paste miRNA sequences
   The first option is to enter miRNA ID (a unique identifier for the miRNA, but it does not have to be a miRBase ID) and miRNA sequence in the textbox. IDs and sequences are should be in csv (comma separated) or fasta formats. miRNA sequence is the ~22 nucleotide of mature miRNA sequence. The input sequence specifies the SNP (or mutation) as [reference allele/derived allele] at the SNP (or mutation) site. Users can enter multiple input entries for multiple miRNAs.
   The second option is to enter either the miRNA IDs (must be the miRBase ID) or dbSNP IDs as input. Users can enter multiple input rows for multiple miRNAs.
   
   *Note: One miRNA could have multiple SNPs in its sequence and for such cases the webserver will give output for each SNP in the miRNA sequence. For a dbSNP id the webserver will only give the result for that SNP. This is visible from our miRNA id example.

   
   
   

Output

1. Enriched functional categories for miRNA target predictions:
   Example:
   

   Enriched functional categories for predicted miRNA target sets

   miRNA ID	Sequence	Reference targets	Derived targets	Reference targets functional enrichment	Derived targets functional enrichment	Common targets functional enrichment
   hsa-miR-593-5p	AGG[C/G]ACCAGCCAGGCAUUGCUCAGC	download	download	display download	display download	display download

   
   Reference targets: Download target prediction results for the input miRNA (with the reference allele).
   Derived targets: Download target prediction results for the input miRNA (with the derived allele).
   Reference targets functional enrichment: Display the enriched functional categories for the reference target genes. Download link can be used for downloading the same content in txt format.
   Derived targets functional enrichment: Display enriched functional categories for derived target genes. Download link can be used for downloading the same content in txt format.
   Common targets functional enrichment: Display the enriched functional categories for the common targets of reference and derived alleles of the miRNA. Download link can be used for downloading the same content in txt format.
   
   *Note: Download all results link can be used for downloading all the results as a zip folder of .txt files.
   
   

2. Functional similarity scores and gene ontology graphs:
   Example:
   

   Functional similarity scores and gene ontology graphs

   miRNA ID	Sequence	Biological process similarity score	Molecular function similarity score	Cellular component similarity score	Gene Ontology figure
   hsa-miR-593-5p	AGG[C/G]ACCAGCCAGGCAUUGCUCAGC	0.551	0.434	0.561	Biological Process Molecular Function Cellular Component

   
   Similarity score: A score (ranging from 0 to 1) for the semantic similarity between the enriched GO terms associated with target gene sets for the reference and derived alleles. A score close to 1 indicates high similarity. A score close to 0 indicates low similarity. "NA" represents no score as there is no significantly enriched GO term.
   Gene Ontology figures: Links to open the Directed Acyclic Graph (DAG) for biological process, molecular function and cellular component.
   
   

miRpair2GO

Input

1. Select prediction method:
   Users can select one of the following miRNA target prediction methods
           TargetScan: predict miRNA targets using TargetScan.
           miRanda: predict miRNA targets using miRanda.
   Or users can combine the output of these two prediction methods
           Union of the predicted target sets from TargetScan and miRanda.
           Intersection of the predicted target sets from TargetScan and miRanda.
   
   
   

2. Specify p-value threshold for functional enrichment analysis The p-value threshold is used to determine the significantly enriched functional categories for each target gene set.

   
   
   

3. Select Gene Ontology hierarchical filtering level Hierarchical filtering is used to group the similar GO terms in the hierarchy of the GO graph. For each group of GO terms, one GO term is selected as the representative term for the group. Three hierarchical filtering options are:
   none: No hierarchical filtering, thus all the enriched GO terms are presented independently.
   moderate: Moderate filtering allows a single parent for each group in the GO hierarchy where all the enriched descendent terms of the parent term are included in the group and then the term with the lowest enrichment p-value is considered as representative term for the group.
   strong: Strong filtering allows multiple parents for the same group. For strong filtering option, the groups in the GO term hierarchies are defined by including the GO terms with common descendent in the same group. Thus the total number of representative terms are expected to be lower by using strong filtering.
   
   

4. Paste miRNA pairs
   Paste miRNA id pairs: Users can enter multiple input rows for multiple miRNA pairs.

   
   
   

Output

1. Enriched functional categories for miRNA target predictions:
   Example:
   

   Enriched functional categories for predicted miRNA target sets

   miRNA ID pair miRNA I,miRNA II	miRNA I targets	miRNA II targets	miRNA I targets functional enrichment	miRNA II targets functional enrichment	Common targets functional enrichment
   hsa-miR-1,hsa-miR-9-5p	download	download	display download	display download	display download

   
   miRNA I targets: Download target prediction results for the input miRNA I.
   miRNA II targets: Download target prediction results for the input miRNA II.
   miRNA I targets functional enrichment: Display the enriched functional categories for miRNA I target genes. The download link can be used for downloading the same content in txt format.
   miRNA II targets functional enrichment: Display the enriched functional categories for miRNA II target genes. The download link can be used for downloading the same content in txt format.
   Common targets functional enrichment: Display the enriched functional categories for the common targets of miRNA I and miRNA II. The download link can be used for downloading the same content in txt format.
   
   

2. Functional similarity scores and gene ontology graphs:
   Example:
   

   Functional similarity scores and gene ontology graphs

   miRNA ID pair miRNA I,miRNA II	Biological process similarity score	Molecular function similarity score	Cellular component similarity score	Gene Ontology figure
   hsa-miR-1,hsa-miR-9-5p	0.419	0.74	0.816	Biological Process Molecular Function Cellular Component

   
   Similarity score: A score (ranging from 0 to 1) for the semantic similarity between the enriched GO terms associated with target gene sets for miRNA I and miRNA II. A score close to 1 indicates high similarity. A score close to 0 indicates low similarity. "NA" represents no score as there is no significantly enriched GO term.
   Gene Ontology figures: Links to open the Directed Acyclic Graph (DAG) for biological process, molecular function and cellular component.
   
   
   *A note about web browser usage: The function of displaying GO figures is not supported by the current version of Internet Explorer. The default threshold for connection timeout in Firefox is 300 seconds. It may take longer time to run the queries containing more multiple mutations. Here is an instruction on how to change the connection timeout setting in Firefox. We suggest to use Google Chrome when accessing miR2GO.