Biology subjects

Nikolaou, C.

Publications and source records attributed to Nikolaou, C..

17 recordsLinked to original sources

bioRxiv2026

Evolutionary origins of protein novelty across an entire yeast subphylum

Novel protein-coding sequences fuel molecular and cellular evolutionary innovations and frequently contribute to species-defining characteristics. They can originate either de novo from previously noncoding sequences or through extreme divergence of already coding ones. How frequently each mechanism occurs and how they shape the structural and functional potential of the resulting proteins remains unclear. Here, we conducted a broad computational investigation of genetic and protein novelty throughout the entire subphylum of Saccharomycotina yeasts. We detected more than 5,000 robust de novo genes across 332 species and compared them to more than 6,000 novel genes resulting from extreme sequence divergence, revealing two quantitatively similar but qualitatively distinct modes of evolution of novelty. A remarkable 40% of de novo proteins are predicted to localize to mitochondria compared to only 20% of divergent, with the latter also being substantially longer and more disordered. A detailed analysis of conservatively predicted tertiary structures of novel proteins shows that ''invention'' of new folds occurs more frequently through de novo emergence. We also illustrate cases of evolutionary ''re-invention'' of existing protein folds from noncoding sequences. Our work deepens our understanding of the origins and importance of novel proteins, opening new directions for further structural and functional characterization.

genomics↗

bioRxiv2026

The genome organizer SMC1A mediates the gene expression response in inflammation by dissociating from nuclear-speckles and redistributing to the nuclear periphery

Nuclear architectural proteins are increasingly recognized as multifunctional regulators whose roles extend well beyond static genome organization. Here we report that SMC1A, a core subunit of the cohesin complex, undergoes a striking redistribution upon inflammatory stimulation in human monocytes, dissociating from nuclear speckles and accumulating at genomic regions enriched for stress-response genes with distinct exon-intron architectural properties. Through integrative analysis of RNA-seq, chromatin organization, and nuclear spatial data, we demonstrate that this redistribution has functional consequences at multiple levels of gene expression. SMC1A dissociation from speckles is accompanied by a reduction in intron retention events -consistent with a transition from a poised, pre-loaded transcriptional state toward active, efficient co-transcriptional processing- and by selective engagement with genes whose exon-intron architecture favors exon definition splicing, shorter nuclear mRNA residence times, and peripheral radial positioning. Genes affected by SMC1A silencing, by contrast, occupy more central nuclear positions and display fundamentally different structural properties, demonstrating that the two modes of SMC1A perturbation -stress-induced redistribution and depletion- are functionally and spatially non-equivalent. These findings suggest a genome compartmentalization in which DNA compositional preferences, gene architecture, radial positioning, and splicing mode converge to define gene sets capable of rapid, precise activation. SMC1A navigates this pre-existing landscape upon inflammatory cues, coordinating transcriptional and post-transcriptional responses simultaneously. We propose that stress-induced redistribution of architectural proteins within a largely invariant nuclear compartmental framework represents a general regulatory mechanism, one whose logic is encoded in the structural organization of the genome itself.

genomics↗

bioRxiv2025

Machine learning can identify orphans that have diverged into the ''twilight zone'' of sequence similarity

Species-specific orphan genes (orphans) lack homologues outside of a given taxon and frequently underlie unique species traits. It is thus important to elucidate the evolutionary origins of orphans. Orphan genes can result from sequence divergence beyond recognition, when homologous proteins diverge to an extent at which tools that rely on sequence similarity to establish homology can no longer identify them as homologues. Orphans can also result from other processes, including de novo gene emergence from previously noncoding sequences, in which a homologous protein-coding gene truly does not exist. Here we propose that orphans resulting from divergence might be recognizable from their patterns of non-statistically significant similarity hits which are almost always discarded. To test this, we simulated diverged orphan protein sequences based on conserved proteins from the Unified Human Gastrointestinal Protein catalogue (UHGP) and used reversed protein sequences as negative data sets. We trained four machine learning classifiers on features extracted from the similarity search tool DIAMONDs output, like total query coverage or maximum bit score. We tested the influence of evolutionary parameters such as simulation tree branch length, indel rate and among-site rate heterogeneity. We found that the performance of the models depends on the simulation parameters: when the underlying simulated divergence was moderate, accuracy reached [~]90%, but when extremely diverged scenarios where simulated accuracy dropped to [~]70%. The most important features for the classification were the number of alignments (hits) and the minimum hit E-value. When applying our classifier on a set of [~]170,000 eligible simulated orphans from the UHGP dataset, we found that [~]30% of them are predicted to be divergent and these are shorter and more disordered than the rest. Our classifiers and pre-processing python scripts are available at https://github.com/emiliostassios/Classification-of-divergent-genes-using-ML and can be readily used as a computationally fast means to obtain a candidate set of diverged orphans from any similarity search output. Therefore, our work allows to study such orphans across the tree of life and in doing so to recover cases of remote homology, get better estimates of the evolutionary age of protein families, detect cases of rapid divergence and more generally better understand how genetic novelty arises.

bioinformatics↗

bioRxiv2025

Notch3 regulates pericyte phenotypic plasticity in colorectal cancer

Pericytes undergo phenotypic alterations that influence cancer progression, yet the molecular mechanisms governing these changes remain poorly understood. Here, we investigated the role of Notch3 signaling in pericyte phenotype and functions in colorectal cancer (CRC). Using lineage tracing approaches, we showed that tumor pericytes originate from normal tissue-resident pericytes, which proliferate inside tumors. In vivo genetic manipulation revealed that Notch3 pathway activation promotes pericyte proliferation, while suppressing contractile protein expression, and leads to increased endothelial cell proliferation. In contrast, Notch3 deletion leads to reduced endothelial proliferation and a significant decrease in. This effect is associated with a shift toward a contractile phenotype. Single-cell RNA sequencing analysis uncovered significant pericyte heterogeneity in both mouse colitis-associated cancer and human CRC. It specifically identified functionally distinct subpopulations characterized by differential Notch3 activity, which supported our in vivo and in vitro findings. Our results establish Notch3 as a key regulator of pericyte phenotypic plasticity in CRC and suggest that targeting this pathway could represent a promising strategy for improving therapeutic outcomes through vascular normalization.

cancer biology↗

bioRxiv2024

Synergistic and antagonistic activities of IRF8 and FOS enhancer pairs during an immune cell fate switch

Cell-fate instructive genes tend to be regulated by large clusters of enhancers. Whether and how individual enhancers within such clusters cooperate in regulating gene expression is poorly understood. We have previously developed a computational method, SEGCOND, that identifies hubs consisting of enhancer clusters and their target genes, termed Putative Transcriptional Condensates (PTCs). Using SEGCOND, we identified PTCs in a CEBPA-induced B-cell to macrophage transdifferentiation system. We found them to be enriched for highly expressed, lineage-restricted genes and to associate with BRD4, a component of transcriptional condensates. Here we performed single and combinatorial deletions of enhancers within two active PTCs after transdifferentiation is induced, harboring IRF8 and FOS. Two enhancers within the IRF8 PTC were found to form a backup mechanism when combined, safeguarding IRF8 expression and transdifferentiation kinetics. Unexpectedly, two individual enhancers within the FOS PTC antagonize each other at Day 1 of transdifferentiation, delaying the conversion of B-cells to macrophages and reducing FOS expression, but cooperate to increase FOS levels in Day 7 induced cells. Our results reveal differentiation stage-specific, complex interactions, between individual enhancers within a cluster.

genomics↗

bioRxiv2024

Senescent cells cluster CTCF on nuclear speckles to sustain their splicing program

Senescence --the endpoint of replicative lifespan for normal cells-- is established via a complex sequence of molecular events. One such event is the dramatic reorganization of CTCF into senescence-induced clusters (SICCs). However, the molecular determinants, genomic consequences, and functional purpose of SICCs remained unknown. Here, we combine functional assays, super-resolution imaging, and 3D genomics with computational modelling to dissect SICC emergence. We establish that the competition between CTCF-bound and non-bound loci dictates clustering propensity. Upon senescence entry, cells repurpose SRRM2 --a key component of nuclear speckles-- and BANF1 --a molecular glue for chromosomes-- to cluster CTCF and rewire genome architecture. This CTCF-centric reorganization in reference to nuclear speckles functionally sustains the senescence splicing program, as SICC disruption fully reverts alternative splicing patterns. We therefore uncover a new paradigm, whereby cells translate changes in nuclear biochemistry into architectural changes directing splicing choices so as to commit to the fate of senescence. GRAPHICAL ABSTRACT O_FIG O_LINKSMALLFIG WIDTH=200 HEIGHT=200 SRC="FIGDIR/small/603680v1_ufig1.gif" ALT="Figure 1"> View larger version (85K): org.highwire.dtl.DTLVardef@cfb561org.highwire.dtl.DTLVardef@8cfdf3org.highwire.dtl.DTLVardef@7485d5org.highwire.dtl.DTLVardef@1fe9051_HPS_FORMAT_FIGEXP M_FIG C_FIG HIGHLIGHTSO_LIHMGB2-bound loci compete with CTCF-bound ones for nuclear speckle association C_LIO_LISenescent cells repurpose SRRM2 and BANF1 to cluster CTCF on speckles C_LIO_LIBANF1 is essential, but not sufficient for CTCF clustering C_LIO_LIThe SRRM2 RNA-binding domain directs CTCF clustering C_LIO_LISICCs rewire chromatin positioning to sustain the senescence splicing program C_LI

cell biology↗

bioRxiv2023

Rapid and synchronous chemical induction of replicative-like senescence via a small molecule inhibitor

Cellular senescence is now acknowledged as a key contributor to organismal ageing and late-life disease. Although popular, the study of senescence in vitro can be complicated by the prolonged and asynchronous timing of cells committing to it and its paracrine effects. To address these issues, we repurposed the small molecule inhibitor inflachromene (ICM) to induce senescence to human primary cells. Within six days of treatment with ICM, senescence hallmarks, including the nuclear eviction of HMGB1 and -B2, are uniformly induced across IMR90 cell populations. By generating and comparing various high throughput datasets from ICM-induced and replicative senescence, we uncovered significant similarity of the two states. Notably though, ICM suppresses the proinflammatory secretome associated with senescence, thus alleviating most paracrine effects. In summary, ICM induces a senescence-like phenotype rapidly and synchronously thereby allowing the study of its core regulatory program without any confounding heterogeneity.

genomics↗

bioRxiv2023

Nucleosomal DNA has topological memory

One fundamental yet elusive aspect of the chromosome architecture is the constrained topolome, which refers to how chromatin elements restrain DNA topology. Nucleosomes stabilise negative DNA supercoils, with most nucleosomes typically restraining a DNA linking number difference ({Delta}Lk) of about -1.26. However, whether this capacity is uniform across the genome is unknown. Here, we calculated the {Delta}Lk restrained by over 4000 nucleosomes in yeast cells. To achieve this, we placed each nucleosome in a circular minichromosome and performed Topo-seq, a novel high-throughput procedure to inspect the topology of circular DNA libraries in a single gel electrophoresis. We found that nucleosomes inherently restrain distinct {Delta}Lk values depending on their genomic origin. Nucleosome DNA topologies differ significantly at gene bodies ({Delta}Lk=-1.29), intergenic regions ({Delta}Lk=-1.23), rDNA genes ({Delta}Lk=-1.24) and telomeric regions ({Delta}Lk=-1.07). Nucleosomes nearby the transcription start and termination sites also exhibit singular DNA topologies. These findings demonstrate that nucleosome DNA topology is imprinted by its native chromatin context and persists even when the nucleosome is relocated. This imprinting contributes to nucleosome functional roles and chromatin folding architectures.

molecular biology↗

bioRxiv2022

Intergenic regions of Saccharomycotina yeasts are enriched in potential to form transmembrane domains

Intergenic genomic regions have essential regulatory and structural roles that impose constraints on their sequences. But regions that do not currently encode proteins, also carry the potential to do so in the future. De novo gene emergence, the evolution of novel genes out of previously non-coding sequences has now been established as a potent force for genomic novelty. Recently, it was shown that intergenic regions in the genome of S. cerevisiae harbor pervasive cryptic potential to, if theoretically translated, form transmembrane domains (TM domains) more frequently than expected by chance, a property that we refer to as TM-forming enrichment. The source and biological relevance of this property is unknown. Here we expand the investigation into the TM-forming potential of intergenic regions to the entire Saccharomycotina budding yeast subphylum, in an effort to explain this property and understand its importance. We find pervasive but variable enrichment in TM-forming potential across the subphylum, regardless of the composition and average size of intergenic regions. This cryptic property is evenly spread across the genome, cannot be explained by the hydrophobic content of the sequence, and does not appear to localize to regions containing regulatory motifs. This TM-forming enrichment specifically, and not the actual TM-forming potential, is associated, across genomes, with more TM domains in evolutionarily young genes. Our findings shed light on this newly discovered feature of yeast genomes and constitute a first step towards understanding its evolutionary importance.

genomics↗

bioRxiv2022

PML controls angiogenesis and EMT to limit metastasis in MDAMB231 cells

Promyelocytic Leukemia Protein (PML) is the core protein of nuclear bodies (NBs) that regulate a large number of cellular processes, including, context dependent, tumor -suppressor and pro-oncogenic effects. PML knockdown (KD) in breast cancer lines MDA-MB-231, but not MCF7, cells showed higher cell proliferation, increased migration properties and prolonged stem cell -like survival in line with gene enrichment results from RNA sequencing analysis. MDA-MB-231 PML KD cells showed an increase of hypoxic and mesenchymal characteristics manifested by higher HIF1a and the EMT-TWIST2 protein levels respectively. Mechanistically, PML loss caused an increase of HIF1a and TWIST2 RNA levels. Interestingly, TWIST2 binds to PML. Moreover, PML directly opposed the action of HIF1a and TWIST2 on VEGFa and CD24 reporters, respectively. Tumor xenografts of MDA-MB-231 PML KD cells showed a higher micro vessel content, grew faster and had higher metastatic ability with a preference for lung. Thus, PML opposes the aggressive cancer phenotype by multiple mechanisms that antagonize the HIF-hypoxia-angiogenic and EMT axis.

cancer biology↗

bioRxiv2022

The Wnt/TCF7L1 transcriptional repressor axis drives primitive endoderm formation by antagonizing naive and formative pluripotency

Early during preimplantation development and in heterogeneous mouse embryonic stem cells (mESC) culture, pluripotent cells are specified towards either the primed epiblast or the primitive endoderm (PE) lineage. Canonical Wnt signaling is crucial for safeguarding naive pluripotency and embryo implantation, yet the role and relevance of canonical Wnt inhibition during early mammalian development remains unknown. Here, we demonstrate that transcriptional repression exerted by Wnt/TCF7L1 promotes PE differentiation of mESCs and in preimplantation inner cell mass. Time-series RNA sequencing and promoter occupancy data reveal that TCF7L1 binds and represses genes encoding essential naive pluripotency factors and indispensable regulators of the formative pluripotency program, including Otx2 and Lef1. Consequently, TCF7L1 promotes pluripotency exit and suppresses epiblast lineage formation, thereby driving cells into PE specification. Conversely, deletion of Tcf7l1 abrogates PE differentiation without restraining epiblast priming. Taken together, our study underscores the importance of transcriptional Wnt inhibition in regulating lineage segregation in ESCs and preimplantation embryo development as well as identifies TCF7L1 as key regulator of this process.

developmental biology↗

bioRxiv2022

Distinct chromosomal niches in the genome of S. cerevisiae provide the background for genomic innovation and shape the fate of gene duplicates

Gene duplication is a major source of genomic innovation in all eukaryotes, with large proportions of genes being the result of either small-scale (SSD) or genome-wide duplication (WGD) events. In the model unicellular eukaryote Saccharomyces cerevisiae, of which nearly one third of the genome corresponds to gene duplicates, the two modes of duplication have been shown to follow different evolutionary fates, with SSD genes being more prone to acquire novel functionalities (neofunctionalization) and WGD more likely to retain different parts of the original ancestral function (subfunctionalization). Having previously described aspects of functional compartmentalization for the genes of S. cerevisiae, in this work we set out to investigate the existence of positional preferences of gene duplicates. We found that SSD and WGD genes are organized in distinct gene clusters that are, furthermore, segregated, occupying different regions, with SSD being more peripheral and WGD more centrally positioned close to centromeric chromatin. Duplicate gene clusters differ from the rest of the genome in terms of gene size and spacing, gene expression variability and regulatory complexity. What is more interesting, some of these properties are also shared by singleton genes residing in duplicate-rich regions in a position-dependent manner. Our analysis further reveals particular chromatin architectures in the promoters of duplicate genes, which are generally longer, with less pronounced nucleosome-free regions, strong structural constraints and a larger number of regulatory elements. Such structural features appear to be important for gene evolution as we find SSD gene-pair co-expression to be strongly associated with the similarity of nucleosome positioning patterns. We propose that specific regions of the yeast genome provide a favourable environment for the generation and maintenance of small-scale gene duplicates. The existence of such genomic "niches" is supported by the enrichment of these regions in singleton genes bearing similarities with gene "relics", remnants of recent duplications that have reverted to single gene status. Our findings provide a valuable framework for the study of genomic innovation and suggest taking into account positional preferences in the study of gene emergence and fixation in experimentally and naturally evolving populations.

genomics↗

bioRxiv2021

The 3D enhancer network of the developing T cell genome is controlled by SATB1

Mechanisms of tissue-specific gene expression regulation via spatial coordination of gene promoters and distal regulatory elements are still poorly understood. We investigated the 3D genome organization of developing murine T cells and identified SATB1, a tissue-specific genome organizer, enriched at the anchors of promoter-enhancer chromatin loops. We assessed the function of SATB1 in T cell chromatin organization and compared it to the conventional genome organizer CTCF. SATB1 builds a more refined layer of genome organization upon a CTCF scaffold. To understand the regulatory implications of SATB1 loopscape structure, we generated Satb1fl/flCd4-Cre+ (Satb1 cKO) conditional knockout animals which suffered from autoimmunity. We aimed to identify molecular mechanisms responsible for the deregulation of the immune system in Satb1 cKO animals. H3K27ac HiChIP and Hi-C experiments indicated that SATB1 primarily mediates promoter-enhancer loops affecting master regulator genes (such as Bcl6), the T cell receptor locus and adhesion molecule genes, collectively being critical for cell lineage specification and immune system homeostasis. Our findings unravel the function of a tissue-specific factor that controls transcription programs, via spatial chromatin arrangements complementary to the chromatin structure imposed by ubiquitously expressed genome organizers.

molecular biology↗

bioRxiv2020

Dissecting Hes-centered transcriptional networks in neural stem cell maintenance and tumorigenesis in Drosophila

Neural stem cells divide during embryogenesis and post embryonic development to generate the entire complement of neurons and glia in the nervous system of vertebrates and invertebrates. Studies of the mechanisms controlling the fine balance between neural stem cells and more differentiated progenitors have shown that in every asymmetric cell division progenitors send a Delta-Notch signal back to their sibling stem cells. Here we show that excessive activation of Notch or overexpression of its direct targets of the Hes family causes stem-cell hyperplasias in the Drosophila larval central nervous system, which can progress to malignant tumours after allografting to adult hosts. We combined transcriptomic data from these hyperplasias with chromatin occupancy data for Dpn, a Hes transcription factor, to identify genes regulated by Hes factors in this process. We show that the Notch/Hes axis represses a cohort of transcription factor genes. These are excluded from the stem cells and promote early differentiation steps, most likely by preventing the reversion of immature progenitors to a stem-cell fate. Our results suggest that Notch signalling sets up a network of mutually repressing stemness and anti-stemness transcription factors, which include Hes proteins and Zfh1, respectively. This mutual repression ensures robust transition to neuronal and glial differentiation and its perturbation can lead to malignant transformation.

developmental biology↗

bioRxiv2020

Extensive fragmentation and re-organization of gene co-expression patterns underlie the progression of Systemic Lupus Erythematosus

Systemic Lupus Erythematosus (SLE) is the prototype of autoimmune diseases, characterized by extensive gene expression perturbations in peripheral blood immune cells. Circumstantial evidence suggests that these perturbations may be due to altered epigenetic profiles and chromatin accessibility but the relationship between transcriptional deregulation and genome organization remains largely unstudied. We developed a genomic approach that leverages patterns of gene coexpression from genome-wide transcriptome profiles in order to identify statistically robust Domains of Co-ordinated gene Expression (DCEs). By implementing this method on gene expression data from a large SLE patient cohort, we identify significant disease-associated alterations in gene co-regulation patterns, which also correlate with the SLE activity status. Low disease activity patient genomes are characterized by extensive fragmentation leading to DCEs of smaller size. High disease activity genomes display excessive spatial redistribution of co-expression domains with expanded and newly-appearing (emerged) DCEs. Fragmentation and redistribution of gene coexpression patterns correlate with SLE-implicated biological pathways and clinically relevant endophenotypes such as kidney involvement. Notably, genes lying at the boundaries of split DCEs of low activity genomes are enriched in the interferon and other SLE susceptibility signatures, suggesting the implication of DCE fragmentation at early disease stages. Interrogation of promoter-enhancer interactions from various immune cell subtypes shows that a significant percentage of nested connections are disrupted by a DCE split or depletion in SLE genomes. Collectively, our results underlining an important role for genome organization in shaping gene expression in SLE, could provide valuable insights into disease pathogenesis and the mechanisms underlying disease flares. SignificanceAlthough widespread gene expression changes have been reported in Systemic Lupus Erythematosus (SLE), attempts to link gene deregulation with genome structure have been lacking. Through a computational framework for the segmentation of gene expression data, we reveal extensive fragmentation and reorganization of gene co-regulation domains in SLE, that correlates with disease activity states. Gene co-expression domains pertaining to biological functions implicated in SLE such as the interferon pathway, are being disrupted in patients, while others associated to severe manifestations such as nephritis, emerge in previously uncorrelated regions of the genome. Our results support extensive genome re-organization underlying aberrant gene expression in SLE, which could assist in the early detection of disease flares in patients that are in remission. Graphical Abstract O_FIG O_LINKSMALLFIG WIDTH=200 HEIGHT=114 SRC="FIGDIR/small/922559v1_ufig1.gif" ALT="Figure 1"> View larger version (29K): org.highwire.dtl.DTLVardef@c9fdfborg.highwire.dtl.DTLVardef@1417da2org.highwire.dtl.DTLVardef@9f229borg.highwire.dtl.DTLVardef@1991569_HPS_FORMAT_FIGEXP M_FIG C_FIG

genomics↗

bioRxiv2019

Monitoring the prolonged TNF stimulation in space and time with topological-functional networks

Genes in linear proximity often share regulatory inputs, expression and evolutionary patterns, even in complex eukaryote genomes with extensive intergenic sequences. Gene regulation, on the other hand, is effected through the co-ordinated activation (or suppression) of genes participating in common biological pathways, which are often transcribed from distant loci. Existing approaches for the study of gene expression focus on the functional aspect, taking positional constraints into account only marginally.\n\nIn this work we propose a novel concept for the study of gene expression, through the combination of topological and functional information into bipartite networks. Starting from genome-wide expression profiles, we define extended chromosomal regions with consistent patterns of differential gene expression and then associate these domains with enriched functional pathways. By analyzing the resulting networks in terms of size, connectivity and modularity we can draw conclusions on the way genome organization may underlie the gene regulation program.\n\nWe implement our approach in a detailed RNASeq profiling of sustained TNF stimulation of mouse synovial fibroblasts. Bipartite network analysis suggests that the cytokine response set by TNF, progresses through two distinct transitions. An early generalization of the inflammatory response, marked by an increase in related functions and high connectivity of corresponding genomic loci, that is followed by a late shutdown of immune functions and the redistribution of expression to developmental and cell adhesion pathways and distinct chromosomal regions.\n\nOur results suggest that the incorporation of topological information may provide additional insights in the underlying topological constraints that are shaping gene expression.

bioinformatics↗

bioRxiv2018

Exploring different sequence representations and classification methods for the prediction of nucleosome positioning

MotivationNucleosomes form the first level of DNA compaction and thus bear a critical role in the overall genome organization. At the same time, they modulate chromatin accessibility and, through a dynamic equilibrium with other DNA-binding proteins, may shape gene expression. A number of large-scale nucleosome positioning maps, obtained for various genomes, has compelled the importance of nucleosomes in the regulation of gene expression and has shown constraints in the relative positions of nucleosomes to be much stronger around regulatory elements (i.e. promoters, splice junctions and enhancers). At the same time, the great majority of nucleosome positions appears to be rather flexible. Various computational methods have in the past been used in order to capture the sequence determinants of nucleosome positioning but, as the extent to which DNA sequence preferences may guide nucleosome occupancy largely varies, this has proved to be rather difficult. In order to focus on highly specific sequence attributes, in this work we have analyzed two well-defined sets of nucleosome-occupied sites (NOS) and nucleosome-free-regions (NFR) from the genome of S. cerevisiae, with the use of textual representations.\n\nResultsWe employed 3 different genomic sequence representations (Hidden Markov Models, Bag-of-Words and N-gram Graphs) combined with a number of machine learning algorithms on the task of classifying genomic sequences as nucleosome-free (NFR) or nucleosome-occupied NOS (to be further amended based on updated results). We found that different approaches that involve the usage of different representations or algorithms can be more or less effective at predicting nucleosome positioning based on the textual data of the underlying genomic sequence. More interestingly, we show that N-gram Graphs, a sequence representation that takes into account both k-mer occurrences and relative positioning at various lengths scales is outperforming other methodologies and may thus be a choice of preference for the analysis of DNA sequences with subtle constraints.

bioinformatics↗