Biology subjects

Ivanovic, S.

Publications and source records attributed to Ivanovic, S..

9 recordsLinked to original sources

bioRxiv2026

Summarizing RNA Structural Ensembles via Maximum Agreement Secondary Structures

Summarizing a collection P of related RNA secondary structures is a key challenge in applications like evolutionary analysis, alternative fold studies and mRNA vaccine design. This requires both clustering the input structures into similar groups and identifying the core structural motifs on which they agree or differ. Existing methods fail by focusing on only one of these goals: clustering methods do not output shared motifs, while consensus methods overlook the structural diversity present in the collection. Here, we introduce the MO_SCPLOWAXIMUMC_SCPLOW AO_SCPLOWGREEMENTC_SCPLOW SO_SCPLOWECONDARYC_SCPLOW SO_SCPLOWTRUCTURESC_SCPLOW (MASS) problem, which seeks the largest set F of structural features present in P that partition the input structures into a user-specified number{tau} of distinct clusters. We prove that MASS is NP-hard and also establish its equivalence to a constrained binary matrix projection problem. We present an exact integer linear program, an exact combinatorial algorithm, and a scalable beam-search heuristic. Using simulations we demonstrate the performance of these exact algorithms and heuristics relative to baseline methods that focus on either clustering or identifying a single consensus tree. On real data, we demonstrate that MASS identifies conserved scaffolds in conformational datasets, reveals conserved structural motifs in different species within RNA families, and recovers shared structural features among synonymous transcripts encoding the same protein. MASS provides a general and interpretable framework for summarizing RNA structural organization.

molecular biology↗

bioRxiv2025

Universal Hyb-Seq kits capture considerable intraspecific variation: Less is more in herbarium-inclusive molecular ecology

Target capture sequencing has enhanced the study of plant evolution and molecular ecology, particularly through the access to degraded DNA from herbarium specimens. Universal "off-the-shelf" kits, such as Angiosperms-353, are cheap and readily available but are considered to expose insufficient variation below the species level, because they are designed to target highly conserved regions. However, this remains to be tested in a direct comparison with customised approaches below the species level. In this study, near-identical genotypes from both herbarium and fresh material of the common dandelion (apomictic lineages in Taraxacum officinale F.H.Wigg.) are characterised with customised and universal approaches of target capture sequencing. An RNA-bait panel was designed to capture (i) highly variable loci normally obtained with a Genotyping-by-Sequencing (GBS) approach customised for dandelions; (ii) custom selected genes with potential for environmental adaptation, likely to harbour intraspecific genetic variation; (iii) conserved exons from universal kits (Angiosperms-353; Compositae-COS). Although exons from universal kits yield considerably less intraspecific genetic variation than both customised approaches, they still provided sufficient genetic variation to discriminate between near-identical genotypes of the same apomictic lineage. Given that universal kits save time, money, and the need for genomic reference data, this approach is recommended to increase the number of samples under budgetary constraints while still capturing considerable levels of intraspecific genetic variation.

genomics↗

bioRxiv2025

Improving Multi-Trait Genomic Prediction Efficiency Through The Incorporation Of Synthetic Traits Selected Based on Co-heritability

Genomic prediction (GP) is an essential tool in the field of plant breeding to accelerate the cultivar development pipeline by predicting the performance of unphenotyped lines. The precision of prediction is constrained by the heritability of the target trait when applying a single-trait genomic prediction model. To overcome this limitation, a multi-trait genomic prediction model leveraging high-heritability secondary traits co-heritable with the target trait can boost predictive ability for the target trait. However, this is practically challenging because it requires additional phenotyping effort and prior knowledge of trait co-heritability. This study aimed to assess the efficiency of multi-trait genomic prediction models powered by secondary traits derived from high-throughput phenotyping data when predicting important leaf functional target traits, i.e., nitrogen (N) content and specific leaf area (SLA) in diverse sorghum accessions. Since these traits can be predicted from hyperspectral reflectance data, there is significant potential for other wavelengths within the existing dataset to meet the criteria needed to improve prediction accuracy using multi-trait approaches. Therefore, experiments were performed on traditional direct measures of leaf N content and SLA, plus partial least squares regression predictions of them (Leaf N-PLSR, SLA-PLSR), i.e., four target traits in total. Three secondary, "synthetic traits" (S1, S2, S3), each a ratio of two wavelengths within the hyperspectral data, were identified based on high co-heritability with a given target trait. Single-trait GBLUP (Genomic Best Linear Unbiased Predictor) was fitted as a baseline model, followed by three multi-trait GBLUP models using synthetic traits and target traits together. Model performance was assessed using k-fold (k=5) cross-validation (CV), which consisted of single-trait, CV1, and CV2 schemes. The synthetic traits high genetic correlation and heritability met the requirements for their use as secondary traits. There was a significant increase in accuracy when synthetic traits were used in the multi-trait genomic prediction model compared to a single trait alone for all four target traits. It improved prediction accuracy while using secondary traits derived from hyperspectral high-throughput phenotyping data in the multi-trait genomic prediction model, suggesting that this approach could be broadly applied in a post-hoc fashion to many datasets without any additional phenotyping effort. Our analysis highlights a practical approach to improve multi-trait genomic prediction model performance using synthetic traits with no intrinsic biological meaning selected through co-heritability estimation.

genetics↗

bioRxiv2024

Evolution-Aware Deep Reinforcement Learning for Single-Cell DNA Copy Number Calling

Recent high-throughput single cell DNA sequencing technologies enable one to detect copy number aberrations (CNAs) on thousands of individual cells within a tumor. Several new methods enable haplotype-specific copy number calling on such data. However, these algorithms do not utilize evolutionary constraints, leading to the inference of spurious CNAs due to measurement noise that are inconsistent with a realistic evolutionary model. To address this gap, we introduce DeepCopy, an evolution-aware deep reinforcement learning algorithm for haplotype-specific copy number calling on single-cell DNA sequencing data. On simulated data, DeepCopys predictions better fit the ground truth than existing methods. As shown on 10x Genomics Single Cell CNV sequencing of several breast cancers and DLP+ sequencing of an ovarian cancer, DeepCopys joint estimation of copy number profiles and evolutionary model parameters results in larger, more realistic clones of cells with identical copy number profiles that lead to more parsimonious CNA phylogenies than existing methods, while retaining agreement with truncal single-nucleotide variants. Finally, on a breast cancer patient sequenced with both ACT and 10x technologies, we demonstrate the consistency of DeepCopys predictions across sequencing technologies.

cancer biology↗

bioRxiv2023

Some like it hot: adaptation to the urban heat island in common dandelion.

The Urban Heat Island Effect (UHIE) is a globally consistent pressure on species living in cities. Rapid adaptation to the UHIE may be necessary for urban wild flora to persist in cities, but experimental evidence is lacking. Here, we report the first evidence of genetic differentiation in a plant species in response to the UHIE. We collected seeds from common dandelion (Taraxacum officinale) individuals along an urban-rural gradient in the city of Amsterdam (The Netherlands). In common-environment greenhouse experiments, we assessed the effect of elevated temperatures on plant growth and the effect of vernalisation treatments on flowering phenology. We found that urban plants accumulate more biomass at higher temperatures and require shorter vernalisation to induce flowering compared to rural plants. Differentiation was also observed between different intra-urban subhabitats, with park plants displaying a higher vernalisation requirement than street plants. Our results show strong differentiation between urban and rural dandelions in temperature-dependent growth and phenology, consistent with adaptive divergence in response to the UHIE. Rapid adaptation to the UHIE may be a potential explanation for the widespread success of dandelions in urban environments. Summary statementThe urban heat island effect (UHIE) is the most prominent and globally consistent characteristic of environmental change due to urbanisation, severely impacting human populations in cities as well as the cohabiting wildlife. Despite the profoundly mitigating effect of vegetation on urban heat, evidence for plant adaptation to the UHIE has been lacking. Here we provide the first experimental evidence to date, demonstrating adaptation in urban dandelions in response to elevated temperatures, similar to the UHIE. We furthermore show an urban-rural differentiation in flowering response to shorter vernalisation times (cold winter period to activate the onset of flowering in early spring). Given the predominantly asexual apomictic mode of reproduction in dandelions, this evolution is likely the result of environmental filtering on a diverse population of clonal genotypes. We conclude that plant adaptation to the UHIE exists and recommend future studies to contrast our findings with those in outcrossing sexual plant systems. Studies of urban heat adaptation can bring impactful contributions to building climate change-resilient environments and plants should be an integral part of this research.

plant biology↗

bioRxiv2023

DNA methylation responses to stress across different plant species

O_LIChanges in environments trigger plant DNA methylation responses, potentially modulating stress responses. Studies on stress-induced DNA methylation typically focus on single species, limiting our understanding of what are general and specific responses between species. C_LIO_LIUsing reduced-representation bisulfite sequencing epiGBS, we compared DNA methylation stress-responses across seven plant species. Because methylation can be targeted at transposable elements (TEs) and because environmental plasticity may be particularly relevant in asexual species, we hypothesize that genome size and reproduction mode explain differences in methylation responses between species. C_LIO_LIWe show that enrichment of environmentally-induced methylation responses in genes and TEs is a general feature across plant species. While previous studies have emphasized methylation responses in CHH-cytosines, we observed that cytosines in all sequence contexts (CG, CHG, CHH) are equally likely to respond to stress. Larger-genome species showed a higher proportion of stress-responding cytosines, and asexual species showed more cytosines with a strong methylation response to stress than sexually responding species. C_LIO_LIOur study highlights the role of TEs in methylome plasticity and identifies causes of variation in methylome plasticity. This provides guidance to extrapolating results from models to other plant species, and may contribute to better understanding of functionality of the response. C_LI

plant biology↗

bioRxiv2022

Modeling and Predicting Cancer Clonal Evolution with Reinforcement Learning

Cancer results from an evolutionary process that typically yields multiple clones with varying sets of mutations within the same tumor. Accurately modeling this process is key to understanding and predicting cancer evolution. Here, we introduce CloMu (Clone To Mutation), a flexible and low-parameter tree-generative model of cancer evolution. CloMu uses a two-layer neural network trained via reinforcement learning to determine the probability of new mutations based on the existing mutations on a clone. CloMu supports several prediction tasks, including the determination of evolutionary trajectories, tree selection, causality and interchangeability between mutations, and mutation fitness. Importantly, previous methods support only some of these tasks, and many suffer from overfitting on datasets with a large number of mutations. Using simulations, we demonstrate that CloMu either matches or outperforms current methods on a wide variety of prediction tasks. In particular, for simulated data with interchangeable mutations, current methods are unable to uncover causal relationships as effectively as CloMu. On breast cancer and leukemia cohorts, we show that CloMu determines similarities and causal relationships between mutations as well as the fitness of mutations. We validate CloMus inferred mutation fitness values for the leukemia cohort by comparing them to clonal proportion data not used during training, showing high concordance. In summary, CloMus low-parameter model facilitates a wide range of prediction tasks regarding cancer evolution on increasingly available cohort-level datasets.

cancer biology↗

bioRxiv2022

DNA methylation in clonal Duckweed lineages (Lemna minor L.) reflects current and historical environmental exposures.

While some DNA methylation variants are transgenerationally stable in plants, DNA methylation modifications that are specifically induced by environmental exposure are typically transient and subject to resetting in germ lines, limiting the potential for transgenerational epigenetics stress memory. Asexual reproduction circumvents germlines, and may be more conducive to long-term memory and inheritance of epigenetic marks. This, however, has been poorly explored. Taking advantage of the rapid clonal reproduction of the common duckweed Lemna minor, we tested the hypothesis that a long-term, transgenerational stress memory from exposure to high temperature can be detected in DNA methylation profiles. Using a reduced representation bisulfite sequencing approach (epiGBS), we show that high temperature stress induces DNA hypermethylation at many cytosines in CG and CHG contexts but not in CHH. In addition, a subset of the temperature responsive CHG cytosines, showed differential DNA methylation between in lineages exposed to 30{degrees}C and 24{degrees}C, 3-12 clonal generations after subsequent culturing in a common environment, demonstrating a memory effect of stress that persists over many clonal generations and that is reflected in DNA methylation. Structural annotation revealed that this memory effect in CHG methylation was enriched in TEs. We argue that the observed epigenetic stress memory is likely caused by stable transgenerational persistence of high temperature-induced DNA methylation variants across multiple clonal generations. To the extent that such epigenetic memory has functional consequences for gene expression and phenotypes, this result suggests potential for long-term modulation of stress responses in asexual plants and vegetatively propagated crops.

evolutionary biology↗

bioRxiv2022

UPP2: Fast and Accurate Alignment Estimation of Datasets with Fragmentary Sequences

MotivationMultiple sequence alignment (MSA) is a basic step in many bioinformatics pipelines. However, achieving highly accurate alignments on large datasets, especially those with sequence length heterogeneity, is a challenging task. UPP (Ultra-large multiple sequence alignment using Phylogeny-aware Profiles) is a method for MSA estimation that builds an ensemble of Hidden Markov Models (eHMM) to represent an estimated alignment on the full length sequences in the input, and then adds the remaining sequences into the alignment using selected HMMs in the ensemble. Although UPP provides good accuracy, it is computationally intensive on large datasets. ResultsWe present UPP2, a direct improvement on UPP. The main advance is a fast technique for selecting HMMs in the ensemble that allows us to achieve the same accuracy as UPP but with greatly reduced runtime. We show UPP2 produces more accurate alignments compared to leading MSA methods on datasets exhibiting substantial sequence length heterogeneity, and is among the most accurate otherwise. Availabilityhttps://github.com/gillichu/sepp Contactwarnow@illinois.edu

bioinformatics↗