LncFinder: an integrated platform for long non-coding RNA identification utilizing sequence intrinsic composition, structural information and physicochemical property

Abstract

Discovering new long non-coding RNAs (lncRNAs) has been a fundamental step in lncRNA-related research. Nowadays, many machine learning-based tools have been developed for lncRNA identification. However, many methods predict lncRNAs using sequence-derived features alone, which tend to display unstable performances on different species. Moreover, the majority of tools cannot be re-trained or tailored by users and neither can the features be customized or integrated to meet researchers' requirements. In this study, features extracted from sequence-intrinsic composition, secondary structure and physicochemical property are comprehensively reviewed and evaluated. An integrated platform named LncFinder is also developed to enhance the performance and promote the research of lncRNA identification. LncFinder includes a novel lncRNA predictor using the heterologous features we designed. Experimental results show that our method outperforms several state-of-the-art tools on multiple species with more robust and satisfactory results. Researchers can additionally employ LncFinder to extract various classic features, build classifier with numerous machine learning algorithms and evaluate classifier performance effectively and efficiently. LncFinder can reveal the properties of lncRNA and mRNA from various perspectives and further inspire lncRNA-protein interaction prediction and lncRNA evolution analysis. It is anticipated that LncFinder can significantly facilitate lncRNA-related research, especially for the poorly explored species. LncFinder is released as R package (https://CRAN.R-project.org/package=LncFinder). A web server (http://bmbl.sdstate.edu/lncfinder/) is also developed to maximize its availability.

lncRNA_Mdeep: an alignment-free predictor for long non-coding RNAs identification by multimodal deep learning

“…To evaluate the performance of lncRNA_Mdeep, we rst investigated the performance of lncRNA_Mdeep with different model architectures on human dataset in 10CV test, and shown the effect of different hyper-parameters in DNNs and CNN, then compared lncRNA_Mdeep with eight existing state-of-the-art methods (i.e., CNCI [14], CPAT [15], PLEK [16], lncRNA-MEDL [17], CPC2 [19], lncRNAnet [20], LncFinder 1 and LncFinder 2 [21]) on human and 11 cross-species datasets in independent test. LncFinder 1 means the LncFinder without secondary structure, and LncFinder 2 means LncFinder with secondary structure.…”

“…Several computational methods are developed for distinguishing lncRNAs from protein-coding transcripts. Existing computation methods can mainly categorized into alignment-based methods [8][9][10][11][12][13] and alignment-free methods [14][15][16][17][18][19][20][21]. The alignment-based methods generally align the transcripts against comprehensive reference protein database to predict lncRNAs, for example, CPC [8] aligned transcripts against UniRef90 dataset [22] using BLSATX [23] tool; lncRNA-ID [11] and lncADeep [13] aligned the transcripts against Pfam dataset [24] using HMMER [25] tool.…”

“…This kind of methods heavily rely on the quality of alignments, which will be in uenced by the performance of multiplesequence alignment tools and the quality of reference databases. Furthermore, the alignment process is extremely time-consuming [15,21]. To avoid the drawback caused by alignment, the alignment-free methods are developed to distinguish lncRNAs from protein-coding transcripts.…”

Elucidating another level of epigenetic regulation in osteoarthritis by identifying functionalcis-acting long non-coding RNAs and their targets in articular cartilage

“…In order to filter out transcripts with unknown coding potential, we integrated two sources of evidence: (1) predictions from the alignment-free Coding Potential Assessment Tool (CPAT, using a logistic regression model built from ORF size, Fickett TESTCODE statistic and hexamer usage bias. LncFinder predicts lncRNAs using heterologous features and machine learn model [20]. Transcripts with coding potential predicted by both tools were removed from the dataset.…”

JEDI: Circular RNA Prediction based on Junction Encoders and Deep Interaction among Splice Sites

“…nRC [11] extracts features from the secondary structures of non-conding RNAs and adopts CNNs framework to classify different types of non-coding RNA. lncFinder [19] integrates both the sequence composition and structural information as features and employs random forests. The learning process can be further optimized to predict different types of lncRNA.…”