Probabilistic Prediction of microRNA

ΆΖ Abstract ΆΖ

Multi-species microRNA and target prediction using machine learning algorithms

  • Human microRNA prediction through a probablistic co-learning model and experimental validation
  • Developement of system: a web server for general multi-species microRNA prediction
  • Improvment of pre- and mature miRNA prediction using kernel mixture model

The goal of this project is to develop an efficient prediction algorithm and an integrated system, and to lay the foundation of functional studies and artificial design for interference system.


A probabilistic co-learning model  implement a general miRNA prediction method identifying close homologs as well as even distant. It combines both sequential and structural characteristics for miRNA genes in a probabilistic framework, and simultaneously decides both whether or not a miRNA gene is and whether a region of mature miRNA is by emitting the signals for cleavage site by Drosha
   Previous approaches detecting only close homologs must be miss novel miRNA genes that lack detectable homology. However, ProMiR, implemented using the probabilistic co-learning model, has a merit to screen not only even distantly homologous miRNA precursors but also close homologs. The ProMiR also provides the predicted position of mature miRNAs (cleavage site by Drosha) on the precursors.

  • Developement of Probabilistic Co-learning Model of Sequence and Structure
  • Human pre- and mature miRNA prediction
  • Validation using RealTime PCR if the candidates are drosha substrates
  • 9 new miRNA identification

Nam, J.-W., Shin, K.R., Han, J., Lee, Y., Kim, V.N. and Zhang, B.-T. (2005) Human microRNA prediction through a probabilistic co-learning model of sequence and structure. Nucleic Acids Res, 33, 3570-3581.

Online service

ProMiR I
ProMiR ver 1.5



ProMiR II is an improved version of the ProMiR for more general prediction of miRNAs. ProMiR II consists of three programs: ProMiR-v, ProMiR-c, and ProMiR-g. ProMiR-v searches both conserved and non-conserved miRNAs in the vicinity of a known miRNA. ProMiR-c predicts both conserved and non-conserved miRNAs in the vicinity of a candidate (70~150 nt)ProMiR-g provides a more general service such as, the prediction of miRNA genes in a long sequence (70 nt~10 kb) of various speceis.

    • Conserved, non-conserved, clustered, non-clustered miRNA prediction
    • Incorpating stem-loop filter and conservation score filter
    • Prediction on Genome Browser annoated with known genes and conservation score
    • unrelated species (such as virus) miRNA prediciton

    J.-W Nam*, J.H.Kim*, S.K.Kim, B.-T. Zhang. ProMiR II: a web server for clutered, nonclustered, conserved, nonconserved miRNA prediction.
    NAR webserver issue, I34:W455-W458, 2006.

    Online service

    ProMiR II


 ΆΖ Reasearch blueprint  ΆΖ

ΆΖ Publications ΆΖ

  • B.-T. Zhang, J.-W. Nam. Supervised learning methods for microRNA studies, Machine Learning for Bioinformatics, chapter 9, John Wiley & Sons, 2007. (in press).
  • J.-G. Joung, K.-B. Hwang, J.-W. Nam, S.-J. Kim, B.-T. Zhang. Discovery of microRNA-mRNA modules via population-based probabilistic learning. Bioinformatics, (in press).
  • J.-W. Nam, I.-H. Lee, K.-B. Hwang, S.-B. Park, B.-T. Zhang. Dinucleotide step parameterization of pre-miRNAs using multi-objective evolutionary algorithms, Lecture Notes in Computer Science, EvoBio 2007, (in press).
  • S.-K. Kim*, J.-W. Nam*, J.-K. Rhee, W.-J. Lee, B.-T. Zhang. miTarget: microRNA target-gene prediction using a Support Vector Machine. BMC Bioinformatics, 7(1):411, 2006. Highlight paper
  • J.-W Nam*, J.H.Kim*, S.K.Kim, B.-T. Zhang. ProMiR II: a web server for clutered, nonclustered, conserved, nonconserved miRNA prediction. NAR webserver issue, I34:W455-W458, 2006.
  • J. Han, Y.T. Kim, K.-H Yeom, J.-W. Nam, I.H. Hur, J.-K. Rhee,  B.-T. Zhang and V.N. Kim. Molecular basis for the recognition and processing of primary microRNA by Drosha.  Cell, In press.
  • V.N. Kim and J.-W. Nam, Genomics of microRNA.  Trends in Genetics, 22(3):165-173, 2006. Most downloaded paper
  • S.K. Kim, J.-W. Nam, W.J. Lee, B.T. Zhang. A Kernel Method for MicroRNA Target Prediction Using Sensible Data and Position-Based Features. IEEE Symposium on Computational Intelligence in Bioinformatics and Computational Biology (CIBCB 2005), pp. 46-52, 2005.
  • J.-W. Nam, K.R. Shin, J.J. Han, Y.T. Lee, V.N. Kim and B.T.Zhang. Human miRNA prediction through probabilistic co-learning of sequence and structure. Nucleic Acids Research, 33(11):3570-3581, 2005. Hottest Paper
  • W.-J. Lee, J.-W. Nam, S.-K. Kim, B.-T. Zhang. Identification of C.elegans MicroRNA Targets Using a Kernel Method.  Genomics and Informatics vol. 3(1):15-23, 2005.
  • J.-W. Nam, W.J. Lee, B.T. ZHANG. Computational Methods for Identification of Human microRNA Precursors.  Lecture Notes in Artificial Intelligence, vol. 3157, pp. 732-741, 2004.
  • J.-W. Nam, Joung, J.-G., Ahn, Y.S. and Zhang, B.-T. Two-Step Genetic Programming for Optimization of RNA Common-Structure,  Lecture Notes in Computer Science, vol. 3005, pp. 73-83, 2004.

Project Title

Probabilistic Prediction of microRNA and Target Gene


The Ministry of Science and Technology of Korea (NRL)

Principal Investigator

Prof.Byoung-Tak Zhang


Jin-Wu Nam

Je-Keun Rhee

Soo-Jin Kim


2004 - 2007

Cooperative Research Institutes

School of Biological Sciences, Seoul National University

Contact Jin-Wu Nam
Phone +82-2-880-5890
Fax +82-2-883-9120