Visualized Computational Predictions of Transcriptional Effects by Intronic Endogenous Retroviruses

Abstract

When endogenous retroviruses (ERVs) or other transposable elements (TEs) insert into an intron, the consequence on gene transcription can range from negligible to a complete ablation of normal transcripts. With the advance of sequencing technology, more and more insertionally polymorphic or private TE insertions are being identified in humans and mice, of which some could have a significant impact on host gene expression. Nevertheless, an efficient and low cost approach to prioritize their potential effect on gene transcription has been lacking. By building a computational model based on artificial neural networks (ANN), we demonstrate the feasibility of using machine-learning approaches to predict the likelihood that intronic ERV insertions will have major effects on gene transcription, focusing on the two ERV families, namely Intracisternal A-type Particle (IAP) and Early Transposon (ETn)/MusD elements, which are responsible for the majority of ERV-induced mutations in mice. We trained the ANN model using properties associated with these ERVs known to cause germ-line mutations (positive cases) and properties associated with likely neutral ERVs of the same families (negative cases), and derived a set of prediction plots that can visualize the likelihood of affecting gene transcription by ERV insertions. Our results show a highly reliable prediction power of our model, and offer a potential approach to computationally screen for other types of TE insertions that may affect gene transcription or even cause disease.

Molecular functions of human endogenous retroviruses in health and disease

“…Genomic copies of HERVs are of particular interest because in addition to functional viral genes, they have multitude of regulatory DNA regions serving as promoters [1][2][3], enhancers [4,5], polyadenylation signals [6,7], insulators [8,9] and binding sites for various nuclear proteins [2,[10][11][12][13]. Many families of HERVs exhibit high transcriptional activity in human tissues [1,[14][15][16][17]. HERVs are believed to be remnants of numerous retroviral infections [18][19][20] that occurred repeatedly during primate evolution [18,21].…”

Human Endogenous Retrovirus-K HML-2 integration within RASGRF2 is associated with intravenous drug abuse and modulates transcription in a cell-line model

“…RASGRF2-int, like the majority of HK2 intronic integrations of the human genome, is antisense compared with RASGRF2 (10). Intronic ERVs in mice are mostly antisense and the majority of them do not disrupt normal gene transcription (26), while a minority of antisense intronic mouse ERVs have been shown to disrupt transcription (26,28). The alternative explanation for the above-found associations is that RASGRF2-int is a proxy of a genetically linked (yet unknown to us) polymorphism of RASGRF2, which bears the true causal effect.…”

“…Our primary hypothesis is that RASGRF2-int is modulating transcription of RASGRF2 ( Fig. 1), as some intronic ERVs have been shown to modulate transcription in mice (12,26,27). RASGRF2-int, like the majority of HK2 intronic integrations of the human genome, is antisense compared with RASGRF2 (10).…”

Peer Review #1 of "A systematic review of the application of machine learning in the detection and classification of transposable elements (v0.1)"

“…ML and DL fields are supported by multiple companies and industry research groups, which anticipated the great benefits that artificial intelligence can contribute to genomics, human health (Eraslan et al, 2019), and major crops. Several papers using ML or DL techniques reported that TEs are associated with many human diseases (Zhang et al, 2013). For example, cancer-related long noncoding RNAs have higher SINE and LINE numbers than cancer-unrelated long noncoding RNAs (Zhang et al, 2018).…”

“…PRISMA flow diagram Publication identifier Year Q1 Q2 Q3 Q4 Reference Publication identifier Year Q1 Q2 Q3 Q4 Reference P1 X X X X (Yu, Yu & Pan, 2017) P19 2013 X X X (Loureiro et al, 2013b) P2 X X X (Schietgat et al, 2018) P20 2014 X X (Ma, Zhang & Wang, 2014) P3 X X X (Arango-López et al, 2017) P21 2010 X X (Dashti & Masoudi-Nejad, 2010) P4 X X X (Loureiro et al, 2013a) P22 2010 X X (Ding, Zhou & Guan, 2010) P5 X X X (Tsafnat et al, 2011) P23 2019 X (Jaiswal & Krishnamachari, 2019) P6 X X X (Zhang et al, 2018) P24 2015 X X X X (Girgis, 2015) P7 X X X (Eraslan et al, 2019) P25 2018 X X X (Nakano et al, 2018a) P8 X X X (Douville et al, 2018) P26 2018 X X X (Zamith Santos et al, 2018) P9 X X (Chen et al, 2018) P27 2009 X (Abrusan et al, 2009) P10 X X X X (Ashlock & Datta, 2012) P28 2019 X X (Su, Gu & Peterson, 2019) P11 X X X (Smith et al, 2017) P29 2017 X X X X (Nakano et al, 2017) P12 X X X X (Kamath, De Jong & Shehu, 2014) P30 2014 X X X (Brayet et al, 2014) P13 X X X (Kim et al, 2016) P31 2013 X (Zamani et al, 2013) P14 X X X (Segal et al, 2018) P32 2019 X (Hubbard et al, 2019) P15 X X X (Rawal & Ramaswamy, 2011) P33 2014 X X (Ryvkin et al, 2014) P16 X X X (Tang et al, 2017) P34 2013 X X X X (Zhang et al, 2013) P17 X X X (Ventola et al, 2017) P35 2019 X X…”