Computational identification of piRNA targets on mouse mRNAs

Analysis of somatic piRNAs in the malaria mosquito Anopheles coluzzii reveals atypical classes of genic small RNAs

Piwi-interacting small RNAs (piRNA) play a key role in controlling the activity of transposable elements (TEs) in the animal germline. In diverse arthropod species, including the pathogen vectors mosquitoes, the piRNA pathway is also active in nongonadal somatic tissues, where its targets and functions are less clear. Here, we studied the features of small RNA production in head and thorax tissues of an uninfected laboratory strain of Anopheles coluzzii focusing on the 24-32-nt-long RNAs. Small RNAs derived from repetitive elements constitute a minor fraction while most small RNAs process from long noncoding RNAs (lncRNAs) and protein-coding gene mRNAs. The majority of small RNAs derived from repetitive elements and lncRNAs exhibited typical piRNAs features. By contrast, majority of protein-coding gene-derived 24-32 nt small RNAs lack the hallmarks of piRNAs and have signatures of nontemplated 3' end tailing. Most of the atypical small RNAs exhibit female-biased expression and originate from mitochondrial and nuclear genes involved in energy metabolism. We also identified atypical genic small RNAs in Anopheles gambiae somatic tissues, which further validates the noncanonical mechanism of their production. We discuss a novel mechanism of small RNA production in mosquito somatic tissues and the possible functional significance of genic small RNAs.

Mammalian piRNA target prediction using a hierarchical attention model

BACKGROUND

Piwi-interacting RNAs (piRNAs) are well established for monitoring and protecting the genome from transposons in germline cells. Recently, numerous studies provided evidence that piRNAs also play important roles in regulating mRNA transcript levels. Despite their significant role in regulating cellular RNA levels, the piRNA targeting rules are not well defined, especially in mammals, which poses obstacles to the elucidation of piRNA function.

RESULTS

Given the complexity and current limitation in understanding the mammalian piRNA targeting rules, we designed a deep learning model by selecting appropriate deep learning sub-networks based on the targeting patterns of piRNA inferred from previous experiments. Additionally, to alleviate the problem of insufficient data, a transfer learning approach was employed. Our model achieves a good discriminatory power (Accuracy: 98.5%) in predicting an independent test dataset. Finally, this model was utilized to predict the targets of all mouse and human piRNAs available in the piRNA database.

CONCLUSIONS

In this research, we developed a deep learning framework that significantly advances the prediction of piRNA targets, overcoming the limitations posed by insufficient data and current incomplete targeting rules. The piRNA target prediction network and results can be downloaded from https://github.com/SofiaTianjiaoZhang/piRNATarget .

Databases and computational methods for the identification of piRNA-related molecules: A survey

Piwi-interacting RNAs (piRNAs) are a class of small non-coding RNAs (ncRNAs) that plays important roles in many biological processes and major cancer diagnosis and treatment, thus becoming a hot research topic. This study aims to provide an in-depth review of computational piRNA-related research, including databases and computational models. Herein, we perform literature analysis and use comparative evaluation methods to summarize and analyze three aspects of computational piRNA-related research: (i) computational models for piRNA-related molecular identification tasks, (ii) computational models for piRNA-disease association prediction tasks, and (iii) computational resources and evaluation metrics for these tasks. This study shows that computational piRNA-related research has significantly progressed, exhibiting promising performance in recent years, whereas they also suffer from the emerging challenges of inconsistent naming systems and the lack of data. Different from other reviews on piRNA-related identification tasks that focus on the organization of datasets and computational methods, we pay more attention to the analysis of computational models, algorithms, and performances that aim to provide valuable references for computational piRNA-related identification tasks. This study will benefit the theoretical development and practical application of piRNAs by better understanding computational models and resources to investigate the biological functions and clinical implications of piRNA.

Dietary Sugar Shifts Mitochondrial Metabolism and Small RNA Biogenesis in Sperm

Aims: Increasing concentrations of dietary sugar results in a linear accumulation of triglycerides in male Drosophila, while inducing a U-shaped obesity response in their offspring. Here, using a combination of proteomics and small RNA (sRNA) sequencing, we aimed at understanding the molecular underpinning in sperm for such plasticity. Results: Proteomic analysis of seminal vesicles revealed that increasing concentrations of dietary sugar resulted in a bell-shaped induction of proteins involved in metabolic/redox regulation. Using stains and in vivo redox reporter flies, this pattern could be explained by changes in sperm production of reactive oxygen species (ROS), more exactly mitochondria-derived H₂O₂. By quenching ROS with the antioxidant N-acetyl cysteine and performing sRNA-seq on sperm, we found that sperm miRNA is increased in response to ROS. Moreover, we found sperm mitosRNA to be increased in high-sugar diet conditions (independent of ROS). Reanalyzing our previously published data revealed a similar global upregulation of human sperm mitosRNA in response to a high-sugar diet, suggesting evolutionary conserved mechanisms. Innovation: This work highlights a fast response to dietary sugar in mitochondria-produced H₂O₂ in Drosophila sperm and identifies redox-sensitive miRNA downstream of this event. Conclusions: Our data support a model where changes in the sperm mitochondria in response to dietary sugar are the primary event, and changes in redox homoeostasis are secondary to mitochondrial ROS production. These data provide multiple candidates for paternal intergenerational metabolic responses as well as potential biomarkers for human male fertility.

PIWI-interacting RNAs in human diseases: databases and computational models

PIWI-interacting RNAs (piRNAs) are short 21-35 nucleotide molecules that comprise the largest class of non-coding RNAs and found in a large diversity of species including yeast, worms, flies, plants and mammals including humans. The most well-understood function of piRNAs is to monitor and protect the genome from transposons particularly in germline cells. Recent data suggest that piRNAs may have additional functions in somatic cells although they are expressed there in far lower abundance. Compared with microRNAs (miRNAs), piRNAs have more limited bioinformatics resources available. This review collates 39 piRNA specific and non-specific databases and bioinformatics resources, describes and compares their utility and attributes and provides an overview of their place in the field. In addition, we review 33 computational models based upon function: piRNA prediction, transposon element and mRNA-related piRNA prediction, cluster prediction, signature detection, target prediction and disease association. Based on the collection of databases and computational models, we identify trends and potential gaps in tool development. We further analyze the breadth and depth of piRNA data available in public sources, their contribution to specific human diseases, particularly in cancer and neurodegenerative conditions, and highlight a few specific piRNAs that appear to be associated with these diseases. This briefing presents the most recent and comprehensive mapping of piRNA bioinformatics resources including databases, models and tools for disease associations to date. Such a mapping should facilitate and stimulate further research on piRNAs.

piRNAs and PIWI Proteins as Diagnostic and Prognostic Markers of Genitourinary Cancers

piRNAs (PIWI-interacting RNAs) are small non-coding RNAs capable of regulation of transposon and gene expression. piRNAs utilise multiple mechanisms to affect gene expression, which makes them potentially more powerful regulators than microRNAs. The mechanisms by which piRNAs regulate transposon and gene expression include DNA methylation, histone modifications, and mRNA degradation. Genitourinary cancers (GC) are a large group of neoplasms that differ by their incidence, clinical course, biology, and prognosis for patients. Regardless of the GC type, metastatic disease remains a key therapeutic challenge, largely affecting patients’ survival rates. Recent studies indicate that piRNAs could serve as potentially useful biomarkers allowing for early cancer detection and therapeutic interventions at the stage of non-advanced tumour, improving patient’s outcomes. Furthermore, studies in prostate cancer show that piRNAs contribute to cancer progression by affecting key oncogenic pathways such as PI3K/AKT. Here, we discuss recent findings on biogenesis, mechanisms of action and the role of piRNAs and the associated PIWI proteins in GC. We also present tools that may be useful for studies on the functioning of piRNAs in cancers.

Predicting sequence and structural features of effective piRNA target binding sites

Piwi-interacting RNA (piRNA) targets are usually identified through base pairing between the piRNA seed region and complementary bases on the target mRNAs, which often results in false predictions. Crosslinking immunoprecipitation (CLIP) study emerges as a promising method that enables accurate identification of PIWI-clade-based targets containing RNA-binding sites. In the present study, we have analyzed the piRNA-target CLIP-seq datasets to uncover the additional characteristic features of piRNA targets. We studied important sequence and structural features using IP⁺ and IP^- set targets that might enhance the accuracy of target site predictions. Analysis has revealed substantial enrichment of AU in target sites as well as in and around the 30 nts upstream and downstream of target sites in IP⁺ set relative to IP^- set that might be contributing to lowering the minimal folding energy of target sites of IP⁺ + set that might be easing the base pairing between piRNA and their targets. We have also found a lower MFE threshold (en) and higher miRanda score for piRNA targets. Interestingly, we have found that majority of the target sites are residing within 3'UTR, suggesting 3'UTR as a preferential target site like that of miRNA targets. Thus, we hypothesize that our findings on additional key features of piRNA target sites might be valuable in identifying the potential targets of piRNA accurately, which will aid in decrypting their functional importance. This article is protected by copyright. All rights reserved.

Identifying piRNA targets on mRNAs in C. elegans using a deep multi-head attention network

Background

Piwi-interacting RNAs (piRNAs) are the small non-coding RNAs (ncRNAs) that silence genomic transposable elements. And researchers found out that piRNA also regulates various endogenous transcripts. However, there is no systematic understanding of the piRNA binding patterns and how piRNA targets genes. While various prediction methods have been developed for other similar ncRNAs (e.g., miRNAs), piRNA holds distinctive characteristics and requires its own computational model for binding target prediction.

Results

Recently, transcriptome-wide piRNA binding events in C. elegans were probed by PRG-1 CLASH experiments. Based on the probed piRNA-messenger RNAs (mRNAs) binding pairs, in this research, we devised the first deep learning architecture based on multi-head attention to computationally identify piRNA targeting mRNA sites. In the devised deep network, the given piRNA and mRNA segment sequences are first one-hot encoded and undergo a combined operation of convolution and squeezing-extraction to unravel motif patterns. And we incorporate a novel multi-head attention sub-network to extract the hidden piRNA binding rules that can simulate the biological piRNA target recognition process. Finally, the true piRNA–mRNA binding pairs are identified by a deep fully connected sub-network. Our model obtains a supreme discriminatory power of AUC \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$=$$\end{document}= 93.3% on an independent test set and successfully extracts the verified binding pattern of a synthetic piRNA. These results demonstrated that the devised model achieves high prediction performance and suggests testable potential biological piRNA binding rules.

Conclusions

In this research, we developed the first deep learning method to identify piRNA targeting sites on C. elegans mRNAs. And the developed deep learning method is demonstrated to be of high accuracy and can provide biological insights into piRNA–mRNA binding patterns. The piRNA binding target identification network can be downloaded from http://cosbi2.ee.ncku.edu.tw/data_download/piRNA_mRNA_binding.

A Review of Discovery Profiling of PIWI-Interacting RNAs and Their Diverse Functions in Metazoans

PIWI-interacting RNAs (piRNAs) are a class of small non-coding RNAs (sncRNAs) that perform crucial biological functions in metazoans and defend against transposable elements (TEs) in germ lines. Recently, ubiquitously expressed piRNAs were discovered in soma and germ lines using small RNA sequencing (sRNA-seq) in humans and animals, providing new insights into the diverse functions of piRNAs. However, the role of piRNAs has not yet been fully elucidated, and sRNA-seq studies continue to reveal different piRNA activities in the genome. In this review, we summarize a set of simplified processes for piRNA analysis in order to provide a useful guide for researchers to perform piRNA research suitable for their study objectives. These processes can help expand the functional research on piRNAs from previously reported sRNA-seq results in metazoans. Ubiquitously expressed piRNAs have been discovered in the soma and germ lines in Annelida, Cnidaria, Echinodermata, Crustacea, Arthropoda, and Mollusca, but they are limited to germ lines in Chordata. The roles of piRNAs in TE silencing, gene expression regulation, epigenetic regulation, embryonic development, immune response, and associated diseases will continue to be discovered via sRNA-seq.

The biogenesis and biological function of PIWI-interacting RNA in cancer

Small non-coding RNAs (ncRNAs) are vital regulators of biological activities, and aberrant levels of small ncRNAs are commonly found in precancerous lesions and cancer. PIWI-interacting RNAs (piRNAs) are a novel type of small ncRNA initially discovered in germ cells that have a specific length (24-31 nucleotides), bind to PIWI proteins, and show 2'-O-methyl modification at the 3'-end. Numerous studies have revealed that piRNAs can play important roles in tumorigenesis via multiple biological regulatory mechanisms, including silencing transcriptional and posttranscriptional gene processes and accelerating multiprotein interactions. piRNAs are emerging players in the malignant transformation of normal cells and participate in the regulation of cancer hallmarks. Most of the specific cancer hallmarks regulated by piRNAs are involved in sustaining proliferative signaling, resistance to cell death or apoptosis, and activation of invasion and metastasis. Additionally, piRNAs have been used as biomarkers for cancer diagnosis and prognosis and have great potential for clinical utility. However, research on the underlying mechanisms of piRNAs in cancer is limited. Here, we systematically reviewed recent advances in the biogenesis and biological functions of piRNAs and relevant bioinformatics databases with the aim of providing insights into cancer diagnosis and clinical applications. We also focused on some cancer hallmarks rarely reported to be related to piRNAs, which can promote in-depth research of piRNAs in molecular biology and facilitate their clinical translation into cancer treatment.

Computational Methods and Online Resources for Identification of piRNA-Related Molecules

piRNAs are a class of small non-coding RNA molecules, which interact with the PIWI family and have many important and diverse biological functions. The present review is aimed to provide guidelines and contribute to piRNA research. We focused on the four types of identification models on piRNA-related molecules, including piRNA, piRNA cluster, piRNA target, and disease-related piRNA. We evaluated the types of tools for the identification of piRNAs based on five aspects: datasets, features, classifiers, performance, and usability. We found the precision of 2lpiRNApred was the highest in datasets of model organisms, piRNN had a better performance of datasets of non-model organisms, and 2L-piRNA had the fastest recognition speed of all tools. In addition, we presented an overview of piRNA databases. The databases were divided into six categories: basic annotation, comprehensive annotation, isoform, cluster, target, and disease. We found that piRNA data of non-model organisms, piRNA target data, and piRNA-disease-associated data should be strengthened. Our review might assist researchers in selecting appropriate tools or datasets for their studies, reveal potential problems and shed light on future bioinformatics studies.

The impact of the RASSF1C and PIWIL1 on DNA methylation: the identification of GMIP as a tumor suppressor

INTRODUCTION

Recently we have identified a novel RASSF1C-PIWIL1-piRNA pathway that promotes lung cancer cell progression and migration. PIWI-like proteins interact with piRNAs to form complexes that regulate gene expression at the transcriptional and translational levels. We have illustrated in previous work that RASSF1C modulates the expression of the PIWIL1-piRNA gene axis, suggesting the hypothesis that the RASSF1C-PIWI-piRNA pathway could potentially contribute to lung cancer stem cell development and progression, in part, through modulation of gene methylation of both oncogenic and tumor suppressor genes. Therefore, we tested this hypothesis using a non-small cell lung cancer (NSCLC) cell model to identify Candidate Differentially Methylated Regions (DMRs) modulated by the RASSF1C-PIWIL1-piRNA pathway.

MATERIALS AND METHODS

We studied the impact of over-expressing RASSF1C and knocking down RASSF1C and PIWIL1 expression on global gene DNA methylation in the NSCLC cell line H1299 using the Reduced Representation Bisulfite Sequencing (RRBS) method.

RESULTS

DMRs were identified by comparing DNA methylation profiles of experimental and control cells. Over-expression of RASSF1C and knocking down RASSF1C and PIWIL1 modulated DNA methylation of genomic regions; and statistically significant candidate genes residing DMR regions in lung cancer cells were identified, including oncogenes and tumor suppressors. One of the hypermethylated genes, Gem Interacting Protein (GMIP), displays tumor suppressor properties. GMIP expression attenuates lung cancer cell migration, and its over-expression is associated with longer survival of lung cancer patients.

CONCLUSIONS

The RASSF1C-PIWI-piRNA pathway modulates key oncogenes and tumor suppressor genes. GMIP is hypermethylated by this pathway and has tumor suppressor properties.

Clues of in vivo nuclear gene regulation by mitochondrial short non-coding RNAs

Gene expression involves multiple processes, from transcription to translation to the mature, functional peptide, and it is regulated at multiple levels. Small RNA molecules are known to bind RNA messengers affecting their fate in the cytoplasm (a process generically termed ‘RNA interference’). Such small regulatory RNAs are well-known to be originated from the nuclear genome, while the role of mitochondrial genome in RNA interference was largely overlooked. However, evidence is growing that mitochondrial DNA does provide the cell a source of interfering RNAs. Small mitochondrial highly transcribed RNAs (smithRNAs) have been proposed to be transcribed from the mitochondrion and predicted to regulate nuclear genes. Here, for the first time, we show in vivo clues of the activity of two smithRNAs in the Manila clam, Ruditapes philippinarum. Moreover, we show that smithRNAs are present and can be annotated in representatives of the three main bilaterian lineages; in some cases, they were already described and assigned to a small RNA category (e.g., piRNAs) given their biogenesis, while in other cases their biogenesis remains unclear. If mitochondria may affect nuclear gene expression through RNA interference, this opens a plethora of new possibilities for them to interact with the nucleus and makes metazoan mitochondrial DNA a much more complex genome than previously thought.

piRBase: a comprehensive database of piRNA sequences

PIWI-interacting RNAs are a class of small RNAs that is most abundantly expressed in animal germline. Substantial research is going on to reveal the functions of piRNAs in the epigenetic and post-transcriptional regulation of transposons and genes. To collect and annotate these data, we developed piRBase, a database assisting piRNA functional study. Since its launch in 2014, piRBase has integrated 264 data sets from 21 organisms, and the number of collected piRNAs has reached 173 million. The latest piRBase release (v2.0, 2018) was more focused on the comprehensive annotation of piRNA sequences, as well as the increasing number of piRNAs. In addition, piRBase release v2.0 also contained the potential information of piRNA targets and disease related piRNA. All datasets in piRBase is free to access, and available for browse, search and bulk downloads at http://www.regulatoryrna.org/database/piRNA/.

Cell detachment rapidly induces changes in noncoding RNA expression in human mesenchymal stromal cells

Aims: To identify differential expression of noncoding RNAs after trypsinization in human mesenchymal stromal cells (hMSCs), focusing on miRNAs, piRNAs and circRNAs. Methods: hMSCs from the bone marrow of three donors were collected for RNA extraction, either lysed directly in monolayer or trypsinized and lysed within 30 min. Total RNA was isolated and sequenced for the evaluation of miRNA and piRNA expression or RNaseR treated and labeled for circRNA array hybridization. RT-qPCR was performed to evaluate the stability of candidate reference genes. Results & conclusions: Alterations in levels of several noncoding RNAs are rapidly induced after trypsinization of hMSCs, affecting critical pathways. This should be carefully considered for a proper experimental design.