Dataset Information

MicroRNA target prediction programs predict many false positives.

ABSTRACT: According to the current view, each microRNA regulates hundreds of genes. Computational tools aim at identifying microRNA targets, usually selecting evolutionarily conserved microRNA binding sites. While the false positive rates have been evaluated for some prediction programs, that information is rarely put forward in studies making use of their predictions. Here, we provide evidence that such predictions are often biologically irrelevant. Focusing on miR-223-guided repression, we observed that it is often smaller than inter-individual variability in gene expression among wild-type mice, suggesting that most predicted targets are functionally insensitive to that microRNA. Furthermore, we found that human haplo-insufficient genes tend to bear the most highly conserved microRNA binding sites. It thus appears that biological functionality of microRNA binding sites depends on the dose-sensitivity of their host gene and that, conversely, it is unlikely that every predicted microRNA target is dose-sensitive enough to be functionally regulated by microRNAs. We also observed that some mRNAs can efficiently titrate microRNAs, providing a reason for microRNA binding site conservation for inefficiently repressed targets. Finally, many conserved microRNA binding sites are conserved in a microRNA-independent fashion: Sequence elements may be conserved for other reasons, while being fortuitously complementary to microRNAs. Collectively, our data suggest that the role of microRNAs in normal and pathological conditions has been overestimated due to the frequent overlooking of false positive rates.

SUBMITTER: Pinzon N

PROVIDER: S-EPMC5287229 | biostudies-literature | 2017 Feb

REPOSITORIES: biostudies-literature

ACCESS DATA

Publications

microRNA target prediction programs predict many false positives.

Pinzón Natalia N Li Blaise B Martinez Laura L Sergeeva Anna A Presumey Jessy J Apparailly Florence F Seitz Hervé H

Genome research 20161115 2

According to the current view, each microRNA regulates hundreds of genes. Computational tools aim at identifying microRNA targets, usually selecting evolutionarily conserved microRNA binding sites. While the false positive rates have been evaluated for some prediction programs, that information is rarely put forward in studies making use of their predictions. Here, we provide evidence that such predictions are often biologically irrelevant. Focusing on miR-223-guided repression, we observed that ...[more]

PMID: 28148562

Similar Datasets

Project description:BackgroundA large number of probabilistic models used in sequence analysis assign non-zero probability values to most input sequences. To decide when a given probability is sufficient the most common way is bayesian binary classification, where the probability of the model characterizing the sequence family of interest is compared to that of an alternative probability model. We can use as alternative model a null model. This is the scoring technique used by sequence analysis tools such as HMMER, SAM and INFERNAL. The most prevalent null models are position-independent residue distributions that include: the uniform distribution, genomic distribution, family-specific distribution and the target sequence distribution. This paper presents a study to evaluate the impact of the choice of a null model in the final result of classifications. In particular, we are interested in minimizing the number of false predictions in a classification. This is a crucial issue to reduce costs of biological validation.ResultsFor all the tests, the target null model presented the lowest number of false positives, when using random sequences as a test. The study was performed in DNA sequences using GC content as the measure of content bias, but the results should be valid also for protein sequences. To broaden the application of the results, the study was performed using randomly generated sequences. Previous studies were performed on aminoacid sequences, using only one probabilistic model (HMM) and on a specific benchmark, and lack more general conclusions about the performance of null models. Finally, a benchmark test with P. falciparum confirmed these results.ConclusionsOf the evaluated models the best suited for classification are the uniform model and the target model. However, the use of the uniform model presents a GC bias that can cause more false positives for candidate sequences with extreme compositional bias, a characteristic not described in previous studies. In these cases the target model is more dependable for biological validation due to its higher specificity.

Dataset Information

MicroRNA target prediction programs predict many false positives.

Publications

microRNA target prediction programs predict many false positives.

Similar Datasets

OmicsDI is part of the ELIXIR infrastructure

Tweets