DOCUMENTATION

About CAMi-Finder

CAMi-Finder is a resource for human cancer miRNA associations developed at Machine Intelligence Unit, Indian Statistical Institute, Calcutta, India. It provides two types of functionalities:
a. Showing all the validated cancer-miRNA relationship with a mouse-click and
b. Predicts novel cancer-miRNA relationships by mining predicted miRNA-target genes( those that are onco/tumor-suppressor using Five target prediction algorithms viz.,
Targetminer
Bandyopadhyay S, Mitra R. TargetMiner: MicroRNA target prediction with systematic identification of tissue specific negative examples.Bioinformatics|2009;25:2625-2631.
TargetScan
Friedman R, et al. Most mammalian mRNAs are conserved targets of microRNAs. Genome Res. 2009;19:92-105.
miRanda
Betel D, Wilson M, Gabow A, Marks DS, Sander C., The microRNA.org resource: targets and expression, Nucleic Acids Res. 2008 Jan; 36(Database Issue): D149-53.
PITA
Kertesz M, et al. The role of site accessibility in microRNA target recognition. Nat. Genet. 2007;39:1278-1284.
TargetSpy
Sturm M, Hackenberg M, Langenberger D, Frishman D.,TargetSpy: a supervised machine learning approach for microRNA target prediction, BMC Bioinformatics. 2010;11:292.
:

Current status of CAMi-Finder

As of December 31, 2010 CAMi-Finder hosts 578 validated cancer-miRNA associations between 234 human miRNAs and 29 cancer types.

Mining of Customized Associations Between Cancer Causing MicroRNAs:

This web-interface is for mining novel cancer associated miRNAs from predicted miRNA-(target onco/Tumor-suppressor gene) database using five target prediction algorithms mentioned above. In this regard a set of 3807 cancer causing genes are extracted from the Cancer Gene Index of National Cancer Institute database. These genes are associated with twenty nine cancer types. Each of these cancer types is associated with several human miRNAs provided by the CAMi-Finder web-server[see This web-interface is for mining novel cancer associated miRNAs from predicted miRNA-(target onco/Tumor-suppressor gene) database using five target prediction algorithms mentioned above. In this regard a set of 3807 cancer causing genes are extracted from the Cancer Gene Index of National Cancer Institute database. These genes are associated with twenty nine cancer types. Each of these cancer types is associated with several human miRNAs observed by mining hundreds of articles from PubMed. Five target prediction algorithms, mentioned above, have been used to extract miRNA-cancer causing gene relqationships. Association Rule Mining technique is then used to identify novel associations between potential cancer causing miRNAs, corresponding to the five prediction databases.

User Instructions:

CAMi-Finder provides privilege to the user to customize the result by choosing the target prediction method and confidence level for mining associations between cancer causing miRNAs. User needs to

a. Select one Target prediction database from the drop down list
b. Select Confidence level [Range: 0.7 to 1]

Output: Extracted rules of the form miRNAm &rarr miRNAn depict the fact that if antecedent miRNAs {miRNAm} are involved in one or more cancer types then there is a high chance for consequent miRNAs {miRNAn} to be associated with the same cancer type. Confidence: Confidence is the conditional probability that is associated with Rules that depicts the fact that the consequent will occur given the occurrence of the antecedent. The tool eliminates any rules with confidence below the user defined confidence level.



Mining of Validated Cancer Associated MicroRNAs:

This web-interface provides a large set of validated cancer-miRNA associations by mining hundreds of articles from PubMed. The users can submit their query by clicking on 'Submit Query' button after selecting a. Cancer type: provides all the validated miRNA-cancer asociations that are related to cancer type. b. miRNA name: Provides all the validated miRNA-cancer associations that are related with the selected miRNA name. c. Both cancer type and miRNA name: Provides validated miRNA-cancer association that is related with the selected cancer type and miRNA name. The web-server provides a runtime guidance to the user to select appropriate miRNA name for each cancer type. It shows all the miRNAs name when no cancer type is selected. When a cancer type is selected by the user, a list of specific miRNAs will be appeared that are associated with only the selected cancer type. When the user clicks on the 'Submit Query' button, a detailed information about the biologically validated cancer-miRNA associations will be provided by the web-server. It provides a global perspective on (i) which miRNAs are dysregulated in which cancer types, (ii) dysregulation pattern (up/down regulation), (iii) chromosomal location of the miRNAs with start and end points, (iv) technique used to measure the expression level, (v) available fold change information, (vi) available p-values, (vii) cancerous samples and cell lines used in the experiment, (viii) non-cancerous samples and cell types, and (ix) pubmed ids.


Submit Your Information:

CAMi-Finder provides an easy submission interface through which scientists can submit new, validated cancer-miRNA associations by filling the fields given in the `New Submission' form. Initial submission will be deposited in a temporary file. A proper verification will be conducted on the newly submitted data and authentic submissions will be inserted into the database by the administrator.


Administrator Login:

An Administrator Login interface with secure authentication protocol is provided through which all the newly submitted and verified data will be inserted and irrelevant data will be removed from the database. An update interface is also provided through whcih existing records can be updated by the administrator.


Back to Main page