FEELnc: a tool for long non-coding RNA annotation and its application to the dog transcriptome

Abstract

Whole transcriptome sequencing (RNA-seq) has become a standard for cataloguing and monitoring RNA populations. One of the main bottlenecks, however, is to correctly identify the different classes of RNAs among the plethora of reconstructed transcripts, particularly those that will be translated (mRNAs) from the class of long non-coding RNAs (lncRNAs). Here, we present FEELnc (FlExible Extraction of LncRNAs), an alignment-free program that accurately annotates lncRNAs based on a Random Forest model trained with general features such as multi k-mer frequencies and relaxed open reading frames. Benchmarking versus five state-of-the-art tools shows that FEELnc achieves similar or better classification performance on GENCODE and NONCODE data sets. The program also provides specific modules that enable the user to fine-tune classification accuracy, to formalize the annotation of lncRNA classes and to identify lncRNAs even in the absence of a training set of non-coding RNAs. We used FEELnc on a real data set comprising 20 canine RNA-seq samples produced by the European LUPA consortium to substantially expand the canine genome annotation to include 10 374 novel lncRNAs and 58 640 mRNA transcripts. FEELnc moves beyond conventional coding potential classifiers by providing a standardized and complete solution for annotating lncRNAs and is freely available at https://github.com/tderrien/FEELnc.

An improved de novo assembly and annotation of the tomato reference genome using single-molecule sequencing, Hi-C proximity ligation and optical maps

“…Mikado transcriptome discussed above was filtered for known protein-coding genes. The remaining transcripts were the basis for identification of lncRNAs using FEELnc pipeline (Wucher et al 2017). In total 5,874 lncRNA genes were predicted with 6,694 alternately spliced isoforms.…”

Transcriptome Analysis for Non-Model Organism: Current Status and Best-Practices

“…Accumulating evidence showed that the protein-coding genes are accounted for only 50% of inal assembled transcriptome data. Mining inal non-redundant transcriptome data via long non-coding RNA identiication tools such as PLEK [90], lncRScan-SVM [91], FEELnc [92] or measuring protein coding potential of transcripts using various tools such as coding potential calculator (CPC) [93], coding potential assessment tool (CPAT) [94], coding-non-coding index (CNCI) [95] provides us more information about the transcriptome landscape of non-model organism.…”

Transcriptome Analysis of Non‐Coding RNAs in Livestock Species: Elucidating the Ambiguity

“…Several methodologies have been described to identify/distinguish lncRNAs from mRNAs and successfully applied to livestock species such as coding potential calculator (CPC) [122], PhyLoCSF [123], coding-non-coding index (CNCI) [124], coding potential assessment tool (CPAT) [125], Predictor of Long non-coding RNAs and mRNAs based on an improved k-mer scheme (PLEK) [126] and Flexible Extraction of LncRNAs (FEELnc) [127], etc. The FEELnc program developed by the functional annotation of animal genome project consortium (FAANG) [128] is recommended as a standardized protocol for lncRNA analyses in animal species.…”

“…The FEELnc program developed by the functional annotation of animal genome project consortium (FAANG) [128] is recommended as a standardized protocol for lncRNA analyses in animal species. In order to distinguish lncRNAs from mRNAs, FEELnc program uses a machine-learning method for estimation of a protein-coding score according to the RNA size, open reading frame coverage and multi k-mer usage [127]. The FEELnc program can derive an automatically computed cut-of so it maximizes the lncRNA prediction sensitivity and speciicity.…”