MOTIVATION: MicroRNAs (miRNAs) are crucial regulators of gene expression, but the precise mechanisms governing their binding to target sites remain unclear. A major contributing factor to this is the lack of unbiased experimental datasets for training accurate prediction models. While recent experimental advances have provided numerous miRNA-target interactions, these are solely positive interactions. Generating negative examples in silico is challenging and prone to introducing biases, such as the miRNA frequency class bias identified in this work. Biases within datasets can compromise model generalization, leading models to learn dataset-specific artifacts rather than true biological patterns. RESULTS: We introduce a novel methodology for negative sample generation that effectively mitigates the miRNA frequency class bias. Using this methodology, we curate several new, extensive datasets and benchmark several state-of-the-art methods on them. We find that a simple convolutional neural network model, retrained on some of these datasets, is able to outperform state-of-the-art methods reaching average precision scores between 0.81 and 0.86 in test datasets. This highlights the potential for leveraging unbiased datasets to achieve improved performance in miRNA binding site prediction. To facilitate further research and lower the barrier to entry for machine learning researchers, we provide an easily accessible Python package, miRBench, for dataset retrieval, sequence encoding, and the execution of state-of-the-art models. AVAILABILITY AND IMPLEMENTATION: The miRBench Python package is accessible at https://github.com/katarinagresova/miRBench/releases/tag/v1.0.1.

Central European Institute of Technology Masaryk University Brno 62500 Czech Republic

Centre for Molecular Medicine and Biobanking University of Malta Msida MSD 2080 Malta

Department of Applied Biomedical Science Faculty of Health Sciences University of Malta Msida MSD 2080 Malta

National Centre for Biomolecular Research Faculty of Science Masaryk University Brno 61137 Czech Republic

Zobrazit více v PubMed

Alexiou P, Maragkakis M, Papadopoulos GL  et al.  Lost in translation: an assessment and perspective for computational microRNA target identification. Bioinformatics  2009;25:3049–55. PubMed

Bartel DP. MicroRNAs: genomics, biogenesis, mechanism, and function. Cell  2004;116:281–97. PubMed

Bernstein E, Kim SY, Carmell MA  et al.  Dicer is essential for mouse development. Nat Genet  2003;35:215–7. PubMed

Broughton JP, Lovci MT, Huang JL  et al.  Pairing beyond the seed supports microRNA targeting specificity. Mol Cell  2016;64:320–33. PubMed PMC

Calin GA, Croce CM.  MicroRNA signatures in human cancers. Nat Rev Cancer  2006;6:857–66. PubMed

Chakraborty C, Sharma AR, Sharma G  et al.  Therapeutic miRNA and siRNA: moving from bench to clinic as next generation medicine. Mol Ther Nucleic Acids  2017;8:132–43. PubMed PMC

Chou C-H, Chang N-W, Shrestha S  et al.  miRTarBase 2016: updates to the experimentally validated miRNA-target interactions database. Nucleic Acids Res  2016;44:D239–47. PubMed PMC

Condrat CE, Thompson DC, Barbu MG  et al.  miRNAs as biomarkers in disease: Latest findings regarding their role in diagnosis and prognosis. PubMed DOI PMC

Dai R, Ahmed SA.  MicroRNA, a new paradigm for understanding immunoregulation, inflammation, and autoimmune diseases. Transl Res  2011;157:163–79. PubMed PMC

Didiano D, Hobert O.  Perfect seed pairing is not a generally reliable predictor for miRNA–target interactions. Nat Struct Mol Biol  2006;13:849–51. PubMed

Esquela-Kerscher A, Slack FJ.  Oncomirs—microRNAs with a role in cancer. Nat Rev Cancer  2006;6:259–69. PubMed

Grešová K, Alexiou P, Giassa I-C  et al.  Small RNA targets: advances in prediction tools and high-throughput profiling. Biology (Basel)  2022;11:1798. PubMed PMC

Grosswendt S, Filipchyk A, Manzano M  et al.  Unambiguous identification of miRNA: target site interactions by different types of ligation reactions. Mol Cell  2014;54:1042–54. PubMed PMC

Guo H, Viktor HL.  Learning from imbalanced data sets with boosting and data generation. SIGKDD Explor  2004;6:30–9.

Hébert SS, De Strooper B.  Alterations of the microRNA network cause neurodegenerative disease. Trends Neurosci  2009;32:199–206. PubMed

Hejret V, Varadarajan NM, Klimentova E  et al.  Analysis of chimeric reads characterises the diverse targetome of AGO2-mediated regulation. Sci Rep  2023;13:22895. PubMed PMC

He K  et al. Deep residual learning for image recognition. arXiv [cs.CV]. 10.48550/arXiv.1512.03385,  2015, preprint: not peer reviewed. DOI

He L, Hannon GJ.  MicroRNAs: small RNAs with a big role in gene regulation. Nat Rev Genet  2004;5:522–31. PubMed

Helwak A, Kudla G, Dudnakova T  et al.  Mapping the human miRNA interactome by CLASH reveals frequent noncanonical binding. Cell  2013;153:654–65. PubMed PMC

Hsu S-D, Lin F-M, Wu W-Y  et al.  miRTarBase: a database curates experimentally validated microRNA–target interactions. Nucleic Acids Res  2011;39:D163–9. PubMed PMC

Ikeda S, Kong SW, Lu J  et al.  Altered microRNA expression in human heart disease. Physiol Genomics  2007;31:367–73. PubMed

Ivey KN, Srivastava D.  MicroRNAs as regulators of differentiation and cell fate decisions. Cell Stem Cell  2010;7:36–41. PubMed

Klimentová E, Hejret V, Krčmář J  et al.  miRBind: a deep learning method for miRNA binding classification. Genes (Basel)  2022;13:2323. PubMed PMC

Lee RC, Feinbaum RL, Ambros V  et al.  The PubMed

Liu J, Carmell MA, Rivas FV  et al.  Argonaute2 is the catalytic engine of mammalian RNAi. Science  2004;305:1437–41. PubMed

Lorenz R, Bernhart SH, Höner Zu Siederdissen C  et al.  ViennaRNA package 2.0. Algorithms Mol Biol  2011;6:26. PubMed PMC

Manakov SA  et al. Scalable and deep profiling of mRNA targets for individual microRNAs with chimeric eCLIP. bioRxiv, 10.1101/2022.02.13.480296,  2022, preprint: not peer reviewed. DOI

McGeary SE, Bisaria N, Pham TM  et al.  MicroRNA 3′-compensatory pairing occurs through two binding modes, with affinity shaped by nucleotide identity and position. Elife  2022;11:e69803. PubMed PMC

McGeary SE, Lin KS, Shi CY  et al.  The biochemical basis of microRNA targeting efficacy. Science  2019;366:eaav1741. PubMed PMC

Min S, Lee B, Yoon S  et al.  TargetNet: functional microRNA target prediction with deep neural networks. Bioinformatics  2022;38:671–7. PubMed

Morita S, Horii T, Kimura M  et al.  One Argonaute family member, Eif2c2 (Ago2), is essential for development and appears not to be involved in DNA methylation. Genomics  2007;89:687–96. PubMed

O’Connell RM, Taganov KD, Boldin MP  et al.  MicroRNA-155 is induced during the macrophage inflammatory response. Proc Natl Acad Sci U S A  2007;104:1604–9. PubMed PMC

Pla A, Zhong X, Rayner S  et al.  miRAW: a deep learning-based approach to predict microRNA targets by analyzing whole microRNA transcripts. PLoS Comput Biol  2018;14:e1006185. PubMed PMC

Rad SMAH, Halpin JC, Tawinwung S  et al.  MicroRNA-mediated metabolic reprogramming of chimeric antigen receptor T cells. Immunol Cell Biol  2022;100:424–39. PubMed PMC

Van Nostrand E, Pratt G  Shishkin A  et al.  Robust transcriptome wide discovery of RNA-binding protein binding sites with enhanced CLIP (eCLIP). Nat Methods  2016;13:508–14. 10.1038/nmeth.3810 PubMed DOI PMC

van Rooij E, Sutherland LB, Liu N  et al.  A signature pattern of stress-responsive microRNAs that can evoke cardiac hypertrophy and heart failure. Proc Natl Acad Sci U S A  2006;103:18255–60. PubMed PMC

van Rooij E, Olson EN.  MicroRNA therapeutics for cardiovascular disease: opportunities and obstacles. Nat Rev Drug Discov  2012;11:860–72. PubMed PMC

Rupaimoole R, Slack FJ.  MicroRNA therapeutics: towards a new era for the management of cancer and other diseases. Nat Rev Drug Discov  2017;16:203–22. PubMed

Shen L, Yang J, Zuo C  et al.  Circular mRNA-based TCR-T offers a safe and effective therapeutic strategy for treatment of cytomegalovirus infection. Mol Ther  2024;32:168–84. PubMed PMC

Sonkoly E, Pivarcsi A.  Advances in microRNAs: implications for immunity and inflammatory diseases. J Cell Mol Med  2009;13:24–38. PubMed PMC

Thum T, Condorelli G.  Long noncoding RNAs and microRNAs in cardiovascular pathophysiology. Circ Res  2015;116:751–62. PubMed

Vlachos IS, Paraskevopoulou MD, Karagkouni D  et al.  DIANA-TarBase v7.0: indexing more than half a million experimentally supported miRNA: mRNA interactions. Nucleic Acids Res  2015;43:D153–9. PubMed PMC

Yang T-H, Chen J-C, Lee Y-H  et al.  Identifying human miRNA target sites via learning the interaction patterns between miRNA and mRNA segments. J Chem Inf Model  2024;64:2445–53. PubMed

Zhao Y, Samal E, Srivastava D  et al.  Serum response factor regulates a muscle-specific microRNA that targets Hand2 during cardiogenesis. Nature  2005;436:214–20. PubMed

Zheng X, Chen L, Li X  et al.  Prediction of miRNA targets by learning from interaction sequences. PLoS One  2020;15:e0232578. PubMed PMC