miRDB is an online resource for microRNA (miRNA) target prediction and functional annotations. Developed by Nathan Wong and Xiaowei Wang, miRDB provides a platform for researchers to identify genes regulated by miRNAs. The latest update includes 2.1 million predicted gene targets regulated by 6709 miRNAs. A new web server interface allows users to submit their own sequences for miRNA target prediction, offering flexibility in studying custom miRNAs or target genes. Additionally, miRDB now includes a FuncMir Collection, which identifies 568 and 452 functional miRNAs in humans and mice, respectively, based on combined computational analysis and literature mining. The miRDB web interface enables users to search for miRNA targets using either miRNA or gene target search terms. Users can also download target prediction results for individual miRNAs or gene targets. The system uses the MirTarget algorithm for target prediction, which has been updated with additional model training data from experimentally identified miRNA-target pairs. miRDB also integrates with the miRBase database, providing dynamic links to specific miRNA target prediction data. The FuncMir Collection helps researchers focus on functionally relevant miRNAs by combining computational analysis with literature mining. Four selection criteria were used to identify functional miRNAs: PubMed literature mining, sequence conservation, expression profile, and miRBase annotations. A composite score was calculated for each miRNA based on these criteria, with a score of 3 or higher indicating functional relevance. This approach identified 568 and 452 functional precursor miRNAs for humans and mice, respectively. These miRNAs, along with their functional annotations, are included in the FuncMir Collection of miRDB. The system also allows users to search for unconventional target sites in the coding region or 5'-UTR. miRDB is hosted at Washington University and is available at http://mirdb.org.miRDB is an online resource for microRNA (miRNA) target prediction and functional annotations. Developed by Nathan Wong and Xiaowei Wang, miRDB provides a platform for researchers to identify genes regulated by miRNAs. The latest update includes 2.1 million predicted gene targets regulated by 6709 miRNAs. A new web server interface allows users to submit their own sequences for miRNA target prediction, offering flexibility in studying custom miRNAs or target genes. Additionally, miRDB now includes a FuncMir Collection, which identifies 568 and 452 functional miRNAs in humans and mice, respectively, based on combined computational analysis and literature mining. The miRDB web interface enables users to search for miRNA targets using either miRNA or gene target search terms. Users can also download target prediction results for individual miRNAs or gene targets. The system uses the MirTarget algorithm for target prediction, which has been updated with additional model training data from experimentally identified miRNA-target pairs. miRDB also integrates with the miRBase database, providing dynamic links to specific miRNA target prediction data. The FuncMir Collection helps researchers focus on functionally relevant miRNAs by combining computational analysis with literature mining. Four selection criteria were used to identify functional miRNAs: PubMed literature mining, sequence conservation, expression profile, and miRBase annotations. A composite score was calculated for each miRNA based on these criteria, with a score of 3 or higher indicating functional relevance. This approach identified 568 and 452 functional precursor miRNAs for humans and mice, respectively. These miRNAs, along with their functional annotations, are included in the FuncMir Collection of miRDB. The system also allows users to search for unconventional target sites in the coding region or 5'-UTR. miRDB is hosted at Washington University and is available at http://mirdb.org.