Biology subjects

Bornelöv, S.

Publications and source records attributed to Bornelöv, S..

6 recordsLinked to original sources

bioRxiv2026

Lifestyles of Gypsy-family transposons shape their regulatory mechanisms

Transposable elements are a highly diverse group of selfish genomic elements, prevalent across the tree of life, whose uncontrolled propagation poses a threat to genome stability. Recent studies have explored the evolution of Drosophila melanogaster transposable elements, their co-evolution with the host genome, and mechanisms that regulate their activity. However, little is known about their cross-species evolutionary patterns. Long terminal repeat (LTR) retrotransposons are the most active group of transposable elements in Drosophila. They are broadly separated into retroelements, which are active in the germline, and insect endogenous retroviruses that express in the gonadal soma. Somatic elements are hypothesised to infect the germline through their acquisition of virus-derived proteins such as Envelope and sORF2, thus multiplying through successive generations. In this study, we curated the sequences of LTR retrotransposons in 249 drosophilid genomes, allowing us to study their evolution across these species and highlight their varying degrees of conservation. Furthermore, we reveal multiple instances of Envelope protein loss or inactivation that suggest shifts in the expression pattern of these transposons, likely accompanied by adopting different transcriptional control mechanisms. We contrast this with the evolutionary history of sORF2, which we found to be much more stable. Lastly, we examined variations in transposon LTR regions responsible for transcriptional regulation and use predictive modelling to suggest six transcription factors likely involved in their tissue-specific expression. Altogether, we reveal complex, interspecies evolutionary patterns of Gypsy-family LTR retrotransposons and highlight examples of their coevolution with their host genome.

genomics↗

bioRxiv2025

Predicting gene expression using millions of yeastpromoters reveals cis-regulatory logic

MotivationGene regulation involves complex interactions between multiple transcription factors. While early attempts to train deep neural networks to predict gene expression were limited to naturally occurring promoter sequences, the advent of gigantic parallel reporter assays has expanded available training data by orders of magnitude. Despite these advances, a clear understanding of how to use deep learning to study gene regulation is still lacking. MethodHere we investigate the complex association between gene promoters and expression in S. cerevisiae using Camformer, a residual convolutional neural network that ranked 4th in the Random Promoter DREAM Challenge 2022. We present the original model trained on 6.7 million random promoter sequences and investigate 270 alternative models to determine what factors contribute most to model performance. Finally, we use explainable AI to uncover regulatory signals. ResultsWe show that Camformer accurately decodes the association between promoters and gene expression (r2 = 0.914 {+/-} 0.003,{rho} = 0.962 {+/-} 0.002) and provides a substantial improvement over previous state of the art. Furthermore, we show that a much smaller model with approximately 90% fewer parameters than the original model can achieve a high predictive performance. Using Grad-CAM and in silico mutagenesis, we demonstrate that the model learns both individual motifs and their hierarchy. For example, while an IME1 motif on its own increases gene expression, the co-occurrence of a UME6 motif provides a switch to strongly reduce gene expression. Thus, deep learning models such as Camformer can provide detailed insights into cis-regulatory logic. Availability and ImplementationThe data and code used and developed in our experiments are publicly available at: https://github.com/Bornelov-lab/Camformer.

bioinformatics↗

bioRxiv2024

FlaHMM: unistrand flamenco-like piRNA cluster prediction in Drosophila species using hidden Markov models

PIWI-interacting RNAs are a class of small non-coding RNAs that are essential for transposon control in animal gonads. In Drosophila ovarian somatic cells, piRNAs are transcribed from large genomic regions called piRNA clusters, which are enriched for transposon fragments and acts as a memory of past invasions. Despite being widely present across Drosophila species, somatic piRNA clusters are notoriously difficult to identify and study due to their lack of sequence conservation and limited synteny. Current identification methods rely either on extensive manual curation or availability of high-throughput small RNA-seq data, limiting large-scale comparative studies. We now present FlaHMM, a hidden Markov model developed to automate genomic annotation of flamenco-like unistrand piRNA clusters in Drosophila species without the need of experimental data beyond a genome assembly. FlaHMM uses transposable element content across 5 or 10 kb bins calculated from genome sequence alone and is thus able to detect candidate piRNA clusters without the need to obtain flies and experimentally perform small RNA sequencing. We show that FlaHMM performs on par with piRNA-guided or manual methods, and thus provides a scalable and efficient approach to piRNA cluster annotation in new genome assemblies. FlaHMM is freely available at https://github.com/Hannon-lab/FlaHMM under an MIT licence.

bioinformatics↗

bioRxiv2024

A dual histone code specifies the binding of heterochromatin protein Rhino to a subset of piRNA source loci

Animal germ cells deploy a specialized small RNA-based silencing system, called the PIWI-interacting RNA (piRNA) pathway, to prevent unwanted expression of transposable elements and maintain genome integrity. In Drosophila germ cells, the majority of piRNA populations originate from dual-strand piRNA clusters, genomic regions highly enriched in transposable element (TE) fragments, via an elaborate protein machinery centred on the heterochromatin protein 1 homolog, Rhino. Although Rhino binds to peptides carrying trimethylated H3K9 in vitro, it is not fully understood why in vivo only a fraction of H3K9me3-decorated heterochromatin is occupied by Rhino. Recent work revealed that Rhino is recruited to a subset of piRNA clusters by the zinc finger protein Kipferl. Here we identify a Kipferl-independent mode of Rhino targeting that, in addition to the previously established role of H3K9me3, also depends on the histone H3 lysine 27 methyltransferase Enhancer of Zeste. At Kipferl-independent sites, we find that Rhino, through its chromodomain, specifically binds to loci marked by both H3K9me3 and H3K27me3. Although the exact mechanism of how Rhino binding is influenced by dual histone modifications remains unclear from a structural and biochemical perspective, our work suggests that combinatorial modifications can play a crucial role in influencing the specificity of chromatin-binding protein interactions. These findings provide an enhanced understanding of the multifaceted mechanisms by which Rhino targets piRNA source loci highlighting the sophisticated epigenetic landscape governing TE silencing in Drosophila germ cells. Our work further reveals a role for dual histone modifications defining the binding specificity of a key chromatin protein.

molecular biology↗

bioRxiv2021

An evolutionarily conserved stop codon enrichment at the 5' ends of mammalian piRNAs

PIWI-interacting RNAs (piRNAs) are small RNAs required to recognize and silence transposable elements. The 5 ends of mature piRNAs are defined through cleavage of long precursor transcripts, primarily by Zucchini (Zuc). Zuc-dependent cleavage typically occurs immediately upstream of a uridine. However, Zuc lacks sequence preference in vitro, pointing towards additional unknown specificity factors. We examined murine piRNAs and revealed a strong and specific enrichment of three sequences (UAA, UAG, UGA)-- corresponding to stop codons--at piRNA 5 ends. Stop codon sequences were also enriched immediately after piRNA processing intermediates, reflecting their Zuc-dependent tail-to-head arrangement. Further analyses revealed that a Zuc in vivo cleavage preference at four sequences (UAA, UAG, UGA, UAC) promotes 5 end stop codons. This observation was conserved across mammals and possibly further. Our work provides new insights into Zuc-dependent cleavage and may point to a previously unrecognized connection between piRNA biogenesis and the translational machinery.

molecular biology↗

bioRxiv2021

Dimerisation of the PICTS complex via LC8/Cut-up drives co-transcriptional transposon silencing in Drosophila

In animal gonads, the PIWI-interacting RNA (piRNA) pathway guards genome integrity in part through the co-transcriptional gene silencing of transposon insertions. In Drosophila ovaries, piRNA-loaded Piwi detects nascent transposon transcripts and instructs heterochromatin formation through the Panoramix-induced co-transcriptional silencing (PICTS) complex, containing Panoramix, Nxf2 and Nxt1. Here, we report that the highly conserved dynein light chain LC8/Cut-up (Ctp) is an essential component of the PICTS complex. Loss of Ctp results in transposon de-repression and a reduction in repressive chromatin marks specifically at transposon loci. In turn, Ctp can enforce transcriptional silencing when artificially recruited to RNA and DNA reporters. We show that Ctp drives dimerisation of the PICTS complex through its interaction with conserved motifs within Panoramix. Artificial dimerisation of Panoramix bypasses the necessity for its interaction with Ctp, demonstrating that conscription of a protein from a ubiquitous cellular machinery has fulfilled a fundamental requirement for a transposon silencing complex.

molecular biology↗