Biology subjects

Gavryushkin, A.

Publications and source records attributed to Gavryushkin, A..

16 recordsLinked to original sources

bioRxiv2025

Accuracy and Scalability of Machine Learning Methods for Genotype-Phenotype Association Data

Many machine learning methods can be applied to predicting phenotypes from genetic data. Which of these methods work best remains an open question, however. To answer this question, we propose to compare a variety of approaches ability to predict a simulated non-linear complex trait. Specifically, we evaluate these methods on their accuracy and scalability with respect to the amount training data available, the noise present in the data, the complexity of the simulated (trait) functions, and their ability to provide insight into the simulated trait. We then compare the best approach to state-of-the-art models in real data, predicting gout in the UK Biobank. We find that transformer encoders outperform all other methods in simulations, and perform comparably to the state-of-the-art with real data, with a promise to scale to significantly larger datasets.

bioinformatics↗

bioRxiv2024

Filtering for highly variable genes and high quality spots improves phylogenetic analysis of cancer spatial transcriptomics Visium data

Phylogenetic relationship of cells within tumours can help us to understand how cancer develops in space and time, iden-tify driver mutations and other evolutionary events that enable can-cer growth and spread. Numerous studies have reconstructed phylo-genies from single-cell DNA-seq data. Here we are looking into the problem of phylogenetic analysis of spatially resolved near single-cell RNA-seq data, which is a cost-efficient alternative (or complemen-tary) data source that integrates multiple sources of evolutionary information including point mutations, copy-number changes, and epimutations. Recent attempts to use such data, although promis-ing, raised many methodological challenges. Here, we explored data-preprocessing and modelling approaches for evolutionary analyses of Visium spatial transcriptomics data. We conclude that using only highly variable genes and accounting for heterogeneous RNA capture across tissue-covered spots improves the reconstructed topological relationships and influences estimated branch lengths.

cancer biology↗

bioRxiv2024

A tractable tree distribution parameterized by clade probabilities and its application to Bayesian phylogenetic point estimation

Bayesian phylogenetic analysis with MCMC algorithms generates an estimate of the posterior distribution of phylogenetic trees in the form of a sample of phylogenetic trees and related parameters. The high dimensionality and non-Euclidean nature of tree space complicates summarizing the central tendency and variance of the posterior distribution in tree space. Here we introduce a new tractable tree distribution and associated point estimator that can be constructed from a posterior sample of trees. Through simulation studies we show that this point estimator performs at least as well and often better than standard methods of producing Bayesian posterior summary trees. We also show that the method of summary that performs best depends on the sample size and dimensionality of the problem in non-trivial ways. Author summaryOur research introduces novel methods to analyse a set of phylogenetic tree topologies, such as those generated by Bayesian Markov Chain Monte Carlo algorithms. We define a new model for a distribution on trees that is based on observed clade frequencies. We study it together with closely related models that are based on observed clade split frequencies. These distributions are easy to work with and, as we show experimentally, provide excellent estimates of the true posterior distribution. Furthermore, we demonstrate that they enable us to find the tree with the highest posterior probability, which acts as a summary tree or point estimate of the distribution. In simulation studies, we show that the new methods performs as least as well or better than existing methods. Additionally, we highlight that choosing the best method for summarizing sets of trees remains challenging, as it depends on the sample size and complexity of the problem in non-trivial ways. This work has the potential to improve the accuracy of phylogenetic studies.

bioinformatics↗

bioRxiv2024

The Fossilised Birth-Death Model is Identifiable

Time-dependent birth-death sampling models have been used in numerous studies for inferring past evolutionary dynamics in different areas, e.g. speciation and extinction rates in macroevolutionary studies, or effective reproductive number in epidemiological studies. These models are branching processes where lineages can bifurcate, die, or be sampled with time-dependent birth, death, and sampling rates, generating phylogenetic trees. It has been shown that in some subclasses of such models, different sets of rates can result in the same distributions of reconstructed phylogenetic trees, and therefore the rates become unidentifiable from the trees regardless of their size. Here we show that widely used time-dependent fossilised birth-death (FBD) models are identifiable. This subclass of models makes more realistic assumptions about the fossilisation process and certain infectious disease transmission processes than the unidentifiable birth-death sampling models. Namely, FBD models assume that sampled lineages stay in the process rather than being immediately removed upon sampling. Identifiability of the time-dependent FBD model justifies using statistical methods that implement this model to infer the underlying temporal diversification or epidemiological dynamics from phylogenetic trees or directly from molecular or other comparative data. We further show that the time-dependent fossilised-birth-death model with an extra parameter, the removal after sampling probability, is unidentifiable. This implies that in scenarios where we do not know how sampling affects lineages we are unable to infer this extra parameter together with birth, death, and sampling rates solely from trees.

evolutionary biology↗

bioRxiv2023

Automated convergence diagnostic for phylogenetic MCMC analyses

Assessing convergence of Markov chain Monte Carlo (MCMC) based analyses is crucial but challenging, especially so in high dimensional and complex spaces such as the space of phylogenetic trees (treespace). In practice, it is assumed that the target distribution is the unique stationary distribution of the MCMC and convergence is achieved when samples appear to be stationary. Here we leverage recent advances in computational geometry of the treespace and introduce a method that combines classical statistical techniques and algorithms with geometric properties of the treespace to automatically evaluate and assess practical convergence of phylogenetic MCMC analyses. Our method monitors convergence across multiple MCMC chains and achieves high accuracy in detecting both practical convergence and convergence issues within treespace. Furthermore, our approach is developed to allow for real-time evaluation during the MCMC algorithm run, eliminating any of the chain post-processing steps that are currently required. Our tool therefore improves reliability and efficiency of MCMC based phylogenetic inference methods and makes analyses easier to reproduce and compare. We demonstrate the efficacy of our diagnostic via a well-calibrated simulation study and provide examples of its performance on real data sets. Although our method performs well in practice, a significant part of the underlying treespace probability theory is still missing, which creates an excellent opportunity for future mathematical research in this area. The open source package for the phylogenetic inference framework BEAST2, called ASM, that implements these methods, making them accessible through a user-friendly GUI, is available from https://github.com/rbouckaert/asm/. The open source Python package, called tetres, that provides an interface for these methods enabling their applications beyond BEAST2 can be accessed at https://github.com/bioDS/tetres/.

bioinformatics↗

bioRxiv2023

Ranked Subtree Prune and Regraft

Phylogenetic trees are a mathematical formalisation of evolutionary histories between organisms, species, genes, cancer cells, etc. For many applications, e.g. when analysing virus transmission trees or cancer evolution, (phylogenetic) time trees are of interest, where branch lengths represent times. Computational methods for reconstructing time trees from (typically molecular) sequence data, for example Bayesian phylogenetic inference using Markov Chain Monte Carlo (MCMC) methods, rely on algorithms that sample the treespace. They employ tree rearrangement operations such as SPR (Subtree Prune and Regraft) and NNI (Nearest Neighbour Interchange) or, in the case of time tree inference, versions of these that take times of internal nodes into account. While the classic SPR tree rearrangement is well-studied, its variants for time trees are less understood, limiting comparative analysis for time tree methods. In this paper we consider a modification of the classical SPR rearrangement on the space of ranked phylogenetic trees, which are trees equipped with a ranking of all internal nodes. This modification results in two novel treespaces, which we propose to study. We begin this study by discussing algorithmic properties of these treespaces, focusing on those relating to the complexity of computing distances under the ranked SPR operations as well as similarities and differences to known tree rearrangement based treespaces. Surprisingly, we show the counterintuitive result that adding leaves to trees can actually decrease their ranked SPR distance, which may have an impact on the results of time tree sampling algorithms given uncertain "rogue taxa".

evolutionary biology↗

bioRxiv2023

Estimating the mean in the space of ranked phylogenetic trees

Reconstructing evolutionary histories of biological entities, such as genes, cells, organisms, populations, and species, from phenotypic and molecular sequencing data is central to many biological, palaeontological, and biomedical disciplines. Typically, due to uncertainties and incompleteness in data, the true evolutionary history (phylogeny) is challenging to estimate. Statistical modelling approaches address this problem by introducing and studying probability distributions over all possible evolutionary histories. In practice, computational methods are deployed to learn those distributions typically by sampling them. This approach, however, is fundamentally challenging as it requires designing and implementing various statistical methods over a space of phylogenetic trees (or treespace). Although the problem of developing statistics over a treespace has received substantial attention in the literature and numerous breakthroughs have been made, it remains largely unsolved. The challenge of solving this problem is two-fold: a treespace has non-trivial often counter-intuitive geometry implying that much of classical Euclidean statistics does not immediately apply; many parametrisations of treespace with promising statistical properties are computationally hard, so they cannot be used in data analyses. As a result, there is no single conventional method for estimating even the most fundamental statistics over any treespace, such as mean and variance, and various heuristics are used in practice. Despite the existence of numerous tree summary methods to approximate means of probability distributions over a treespace based on its geometry, and the theoretical promise of this idea, none of the attempts resulted in a practical method for summarising tree samples. In this paper we present such a method along with useful properties of our chosen treespace while focusing on its impact on phylogenetic analyses of real datasets. We perform an extensive benchmark study and demonstrate that our method outperforms currently most popular methods with respect to a number of important "quality" statistics. Further, we apply our method to three real datasets ranging from cancer evolution to linguistics and find novel insights into corresponding evolutionary problems in all of them. We hence conclude that this treespace is a promising candidate to serve as a foundation for developing statistics over phylogenetic trees analytically, as well as new computational tools for evolutionary data analyses.

bioinformatics↗

bioRxiv2022

SNVformer: An Attention-based Deep Neural Network for GWAS Data

Despite being the widely-used gold standard for linking common genetic variations to phenotypes and disease, genome-wide association studies (GWAS) suffer major limitations, partially attributable to the reliance on simple, typically linear, models of genetic effects. More elaborate methods, such as epistasis-aware models, typically struggle with the scale of GWAS data. In this paper, we build on recent advances in neural networks employing Transformer-based architectures to enable such models at a large scale. As a first step towards replacing linear GWAS with a more expressive approximation, we demonstrate prediction of gout, a painful form of inflammatory arthritis arising when monosodium urate crystals form in the joints under high serum urate conditions, from Single Nucleotide Variants (SNVs) using a scalable (long input) variant of the Transformer architecture. Furthermore, we show that sparse SNVs can be efficiently used by these Transformer-based networks without expanding them to a full genome. By appropriately encoding SNVs, we are able to achieve competitive initial performance, with an AUROC of 83% when classifying a balanced test set using genotype and demographic information. Moreover, the confidence with which the network makes its prediction is a good indication of the prediction accuracy. Our results indicate a number of opportunities for extension, enabling full genome-scale data analysis using more complex and accurate genotype-phenotype association models.

bioinformatics↗

bioRxiv2021

A Fast Lasso-Based Method for Inferring Higher-Order Interactions

AO_SCPLOWBSTRACTC_SCPLOWLarge-scale genotype-phenotype screens provide a wealth of data for identifying molecular alterations associated with a phenotype. Epistatic effects play an important role in such association studies. For example, siRNA perturbation screens can be used to identify combinatorial gene-silencing effects. In bacteria, epistasis has practical consequences in determining antimicrobial resistance as the genetic background of a strain plays an important role in determining resistance. Recently developed tools scale to human exome-wide screens for pairwise interactions, but none to date have included the possibility of three-way interactions. Expanding upon recent state-of-the art methods, we make a number of improvements to the performance on large-scale data, making consideration of three-way interactions possible. We demonstrate our proposed method, Pint, on both simulated and real data sets, including antibiotic resistance testing and siRNA perturbation screens. Pint outperforms known methods in simulated data, and identifies a number of biologically plausible gene effects in both the antibiotic and siRNA models. For example, we have identified a combination of known tumor suppressor genes that is predicted (using Pint) to cause a significant increase in cell proliferation. AO_SCPLOWUTHORC_SCPLOW SO_SCPLOWUMMARYC_SCPLOWIn recent years, large-scale genetic datasets have become available for analysis. These large datasets often stretch the limits of classic computational methods, requiring too much memory or simply taking a prohibitively long time to run. Due to the enormous number of potential interactions, each gene or variation in the data is often modeled on its own, without considering interactions between them. Recently, methods have been developed to solve regression problems that include these interacting effects. Even the fastest of these cannot include threeway interactions, however. We improve upon one such method, developing an approach that is significantly faster than the current state of the art. Moreover, our method scales to three-way interactions among thousands of genes, while avoiding a number of the limitations of previous approaches. We analyse large-scale simulated data, antibiotic resistance, and gene-silencing datasets to demonstrate the accuracy and performance of our approach.

bioinformatics↗

bioRxiv2021

A Fast Lasso-based Method for Inferring Pairwise Interactions

AO_SCPLOWBSTRACTC_SCPLOWLarge-scale genotype-phenotype screens provide a wealth of data for identifying molecular alternations associated with a phenotype. Epistatic effects play an important role in such association studies. For example, siRNA perturbation screens can be used to identify pairwise gene-silencing effects. In bacteria, epistasis has practical consequences in determining antimicrobial resistance as the genetic background of a strain plays an important role in determining resistance. Existing computational tools which account for epistasis do not scale to human exome-wide screens and struggle with genetically diverse bacterial species such as Pseudomonas aeruginosa. Combining earlier work in interaction detection with recent advances in integer compression, we present a method for epistatic interaction detection on sparse (human) exome-scale data, and an R implementation in the package Pint. Our method takes advantage of sparsity in the input data and recent progress in integer compression to perform lasso-penalised linear regression on all pairwise combinations of the input, estimating up to 200 million potential effects, including epistatic interactions. Hence the human exome is within the reach of our method, assuming one parameter per gene and one parameter per epistatic effect for every pair of genes. We demonstrate Pint on both simulated and real data sets, including antibiotic resistance testing and siRNA perturbation screens.

bioinformatics↗

bioRxiv2021

Cancer phylogenetics using single-cell RNA-seq data

Phylogenetic methods are emerging as a useful tool to understand cancer evolutionary dynamics, including tumor structure, heterogeneity, and progression. Most currently used approaches utilize either bulk whole genome sequencing (WGS) or single-cell DNA sequencing (scDNA-seq) and are based on calling copy number alterations and single nucleotide variants (SNVs). scRNA-seq is commonly applied to explore differential gene expression of cancer cells throughout tumor progression. The method exacerbates the single-cell sequencing problem of low yield per cell with uneven expression levels. This accounts for low and uneven sequencing coverage and makes SNV detection and phylogenetic analysis challenging. In this paper, we demonstrate for the first time that scRNA-seq data contains sufficient evolutionary signal and can also be utilized in phylogenetic analyses. We explore and compare results of such analyses based on both expression levels and SNVs called from scRNA-seq data. Both techniques are shown to be useful for reconstructing phylogenetic relationships between cells, reflecting the clonal composition of a tumor. Both standardized expression values and SNVs appear to be equally capable of reconstructing a similar pattern of phylogenetic relationship. This pattern is stable even when phylogenetic uncertainty is taken in account. Our results open up a new direction of somatic phylogenetics based on scRNA-seq data. Further research is required to refine and improve these approaches to capture the full picture of somatic evolutionary dynamics in cancer.

cancer biology↗

bioRxiv2020

Learning epistatic gene interactions from perturbation screens

AO_SCPLOWBSTRACTC_SCPLOWThe treatment of complex diseases often relies on combinatorial therapy, a strategy where drugs are used to target multiple genes simultaneously. Promising candidate genes for combinatorial perturbation often constitute epistatic genes, i.e., genes which contribute to a phenotype in a non-linear fashion. Experimental identification of the full landscape of genetic interactions by perturbing all gene combinations is prohibitive due to the exponential growth of testable hypotheses. Here we present a model for the inference of pairwise epistatic, including synthetic lethal, gene interactions from siRNA-based perturbation screens. The model exploits the combinatorial nature of siRNA-based screens resulting from the high numbers of sequence-dependent off-target effects, where each siRNA apart from its intended target knocks down hundreds of additional genes. We show that conditional and marginal epistasis can be estimated as interaction coefficients of regression models on perturbation data. We compare two methods, namely glinternet and xyz, for selecting non-zero effects in high dimensions as components of the model, and make recommendations for the appropriate use of each. For data simulated from real RNAi screening libraries, we show that glinternet successfully identifies epistatic gene pairs with high accuracy across a wide range of relevant parameters for the signal-to-noise ratio of observed phenotypes, the effect size of epistasis and the number of observations per double knockdown. xyz is also able to identify interactions from lower dimensional data sets (fewer genes), but is less accurate for many dimensions. Higher accuracy of glinternet, however, comes at the cost of longer running time compared to xyz. The general model is widely applicable and allows mining the wealth of publicly available RNAi screening data for the estimation of epistatic interactions between genes. As a proof of concept, we apply the model to search for interactions, and potential targets for treatment, among previously published sets of siRNA perturbation screens on various pathogens. The identified interactions include both known epistatic interactions as well as novel findings.

bioinformatics↗

bioRxiv2019

Geometry of Ranked Nearest Neighbour Interchange Space of Phylogenetic Trees

AO_SCPLOWBSTRACTC_SCPLOWIn this paper we study the graph of ranked phylogenetic trees where the adjacency relation is given by a local rearrangement of the tree structure. Our work is motivated by tree inference algorithms, such as maximum likelihood and Markov Chain Monte Carlo methods, where the geometry of the search space plays a central role for efficiency and practicality of optimisation and sampling. We hence focus on understanding the geometry of the space (graph) of ranked trees, the so-called ranked nearest neighbour interchange (RNNI) graph. We find the radius and diameter of the space exactly, improving the best previously known estimates. Since the RNNI graph is a generalisation of the classical nearest neighbour interchange (NNI) graph to ranked phylogenetic trees, we compare geometric and algorithmic properties of the two graphs. Surprisingly, we discover that both geometric and algorithmic properties of RNNI and NNI are quite different. For example, we establish convexity of certain natural subspaces in RNNI which are not convex is NNI. Our results suggest that the complexity of computing distances in the two graphs is different.

bioinformatics↗

bioRxiv2017

High-dimensional microbiome interactions shape host fitness

Gut bacteria can affect key aspects of host fitness, such as development, fecundity, and lifespan, while the host in turn shapes the gut microbiome. Microbiomes co-evolve with their hosts and have been implicated in host speciation. However, it is unclear to what extent individual species versus community interactions within the microbiome are linked to host fitness. Here we combinatorially dissect the natural microbiome of Drosophila melanogaster and reveal that interactions between bacteria shape host fitness through life history tradeoffs. We find that the same microbial interactions that shape host fitness also shape microbiome abundances, suggesting a potential evolutionary mechanism by which microbiome communities (rather than just individual species) may be intertwined in co-selection with their hosts. Empirically, we made germ-free flies colonized with each possible combination of the five core species of fly gut bacteria. We measured the resulting bacterial community abundances and fly fitness traits including development, reproduction, and lifespan. The fly gut promoted bacterial diversity, which in turn accelerated development, reproduction, and aging: flies that reproduced more died sooner. From these measurements we calculated the impact of bacterial interactions on fly fitness by adapting the mathematics of genetic epistasis to the microbiome. Host physiology phenotypes were highly dependent on interactions between bacterial species. Higher-order interactions (involving 3, 4, and 5 species) were widely prevalent and impacted both host physiology and the maintenance of gut diversity. The parallel impacts of bacterial interactions on the microbiome and on host fitness suggest that microbiome interactions may be key drivers of evolution.\n\nSignificanceAll animals have associated microbial communities called microbiomes that can influence the physiology and fitness of their host. It is unclear to what extent individual microbial species versus ecology of the microbiome influences fitness of the host. Here we mapped all the possible interactions between individual species of bacteria with each other and with the hosts physiology. Our approach revealed that the same bacterial interactions that shape microbiome abundances also shape host fitness traits. This relationship provides a feedback that may favor the emergence of co-evolving microbiome-host units.

microbiology↗

bioRxiv2017

The geometry of partial fitness orders and an efficient method for detecting genetic interactions

We present an efficient computational approach for detecting genetic interactions from fitness comparison data together with a geometric interpretation using polyhedral cones associated to partial orderings. Genetic interactions are defined by linear forms with integer coefficients in the fitness variables assigned to genotypes. These forms generalize several popular approaches to study interactions, including Fourier-Walsh coefficients, interaction coordinates, and circuits. We assume that fitness measurements come with high uncertainty or are even unavailable, as is the case for many empirical studies, and derive interactions only from comparisons of genotypes with respect to their fitness, i.e. from partial fitness orders. We present a characterization of the class of partial fitness orders that imply interactions, using a graph-theoretic approach. Our characterization then yields an efficient algorithm for testing the condition when certain genetic interactions, such as sign epistasis, are implied. This provides an exponential improvement of the best previously known method. We also present a geometric interpretation of our characterization, which provides the basis for statistical analysis of partial fitness orders and genetic interactions.

evolutionary biology↗

bioRxiv2017

Inferring Genetic Interactions From Comparative Fitness Data

AO_SCPLOWBSTRACTC_SCPLOWDarwinian fitness is a central concept in evolutionary biology. In practice, however, it is hardly possible to measure fitness for all genotypes in a natural population. Here, we present quantitative tools to make inferences about epistatic gene interactions when the fitness landscape is only incompletely determined due to imprecise measurements or missing observations. We demonstrate that genetic interactions can often be inferred from fitness rank orders, where all genotypes are ordered according to fitness, and even from partial fitness orders. We provide a complete characterization of rank orders that imply higher order epistasis. Our theory applies to all common types of gene interactions and facilitates comprehensive investigations of diverse genetic interactions. We analyzed various genetic systems comprising HIV-1, the malaria-causing parasite Plasmodium vivax, the fungus Aspergillus niger, and the TEM-family of {beta}-lactamase associated with antibiotic resistance. For all systems, our approach revealed higher order interactions among mutations.

evolutionary biology↗