Biology subjects

Chu, G.

Publications and source records attributed to Chu, G..

8 recordsLinked to original sources

bioRxiv2026

LAML-Pro: Maximum Likelihood Inference of Cell Genotypes and Cell Lineage Trees

MotivationRecent dynamic lineage tracing technologies use genome editing to induce heritable mutations, or edits, that accumulate across successive cell divisions. These edits are measured using single-cell sequencing or imaging, providing data to reconstruct cell lineages at single-cell resolution. Current computational approaches to infer cell lineage trees, or phylogenies, from these data perform two separate steps: (1) Identify each cells edits (genotype) from the raw sequencing or imaging data; (2) Infer a cell lineage tree from the cell genotypes. However, genotyping cells is an inexact process and genotype errors can yield an inaccurate lineage tree. For example, using fluorescence based-imaging to measure edits results in a high fraction ({approx} 25-50%) of uncertain or erroneous genotypes. ResultsWe introduce Lineage Analysis via Maximum Likelihood with PRobabilistic Observations (LAML-Pro), an algorithm that jointly infers cell genotypes and a cell lineage tree. LAML-Pro is based on the Probabilistic Mixed-type Missing Observation (PMMO) model, which we derive to describe both the genome editing and genotype observation processes. LAML-Pro constructs lineage trees from thousands of cells in under an hour by leveraging the sparsity of transitions under the PMMO model. On simulated data, we demonstrate that LAML-Pro corrects genotype errors and infers substantially more accurate trees than existing methods which are vulnerable to genotype errors. Applied to data from two recent imaging-based lineage tracing systems, LAML-Pro reduces genotype errors by 5-fold and produces more spatially coherent lineage trees compared to existing methods. Availability and ImplementationLAML-Pro is freely available at: github.com/raphael-group/LAML-Pro.

bioinformatics↗

bioRxiv2025

Cooperation Between Genotoxic Bacteria Accelerates Tumorigenesis in a Mouse Model of Colon Carcinogenesis

To identify causal links between gut microbes and tumorigenesis, we colonized groups of germ-free colon tumor-susceptible mice (ApcMin/+;Il10-/-) with 15 cultured human fecal microbiotas from healthy individuals, as well as patients with inflammatory bowel disease and colorectal cancer. The number of colonic tumors in ApcMin/+;Il10-/- mice varied by donor microbiota but not by the health status of the donor. In vitro screens of host cell proliferation, genotoxicity, and inflammation in bacteria-mammalian cell cocultures revealed that genotoxicity best predicted tumorigenic microbes in vivo with genotoxic microbes present in all tested individuals. The genotoxic subset of strains from each donor induced more tumors than the complete community - even when the complete community was not tumorigenic. Combining genotoxic microbes from multiple sources increased tumor number and decreased the time to tumor onset. Together these results suggest that most individuals harbor genotoxic bacterial strains and the balance of genotoxic to protective strains determines the timing and severity of tumorigenesis in vivo.

microbiology↗

bioRxiv2025

Micromorphology and ammonium transmembrane transport characteristics in roots of rice seedlings

During the seedling stage, rice root exhibit rapid growth and absorb significant quantities of nutrients. This study involved cultivating rice seedlings in two different seedling-raising mediums characterized by high ammonium concentration (HN, NH4+ 1.2 mg g-1) and low ammonium concentration (LN, NH4+ 0.006 mg g-1). The morphology of distinct root zones (root cap, meristematic zone, elongation zone, and maturation zone) was observed under a microscope throughout root development, and variations in transmembrane flux rates of NH4+ in different root zones were assessed using Non-invasive micro-test technology (NMT). Results showed that the root caps of both seminal and adventitious roots became separated from the root tip between 9 and 12 days after root emergence. During this time, the root tip transformed from a sharp to a round shape, and the distance from the maturation zone to the root tip progressively decreased from over 900 m on 6 days after sowing (DAS) to under 500 m on DAS12. The meristematic zone was the primary site for NH4+ absorption in new root and proved to be particularly sensitive to environmental NH4+ concentration. At DAS 6, the net NH4+ flux rate was highest in the meristematic zone under both HN and LN treatments. Additionally, the external NH4+ concentration influenced the direction of NH4+ flux in the meristematic zone at DAS9, with HN being a net NH4+ influx and LN being an NH4+ efflux, and the net NH4+ influx in the meristematic zone persisted for 3 days longer in the HN treatment compared to the LN treatment. In mature roots, the root hair zone emerged as the primary site of NH4+ uptake, exhibiting an infNH4+ influx rate of 40-140 pmol cm-1 s-1, and this low rate of uptake could be sustained for up to 12 days after root emergence. By DAS15, a net NH4+ efflux was observed in the entire seminal root, signaling the loss of NH4+ absorption function 15 days post-emergence. A similar trend was noted in adventitious roots, where NH4+ uptake remained functional for 12-15 days after root emergence. As rice seedlings continued to grow, new adventitious roots replaced old ones, facilitating ongoing NH4+ uptake.

plant biology↗

bioRxiv2024

Maximum Likelihood Inference of Time-scaled Cell Lineage Trees with Mixed-type Missing Data

Recent dynamic lineage tracing technologies combine CRISPR-based genome editing with single-cell sequencing to track cell divisions during development. A key computational problem in dynamic lineage tracing is to infer a cell lineage tree from the measured CRISPR-induced mutations. Three features of dynamic lineage tracing data distinguish this problem from standard phylogenetic tree inference. First, the CRISPR-editing process modifies a genomic location exactly once. This non-modifiable property is not well described by the time-reversible models commonly used in phylogenetics. Second, as a consequence of non-modifiability, the number of mutations per time unit decreases over time. Third, CRISPR-based genome-editing and single-cell sequencing results in high rates of both heritable and non-heritable (dropout) missing data. To model these features, we introduce the Probabilistic Mixed-type Missing (PMM) model. We describe an algorithm, LAML (Lineage Analysis via Maximum Likelihood), to search for the maximum likelihood (ML) tree under the PMM model. LAML combines an Expectation Maximization (EM) algorithm with a heuristic tree search to jointly estimate tree topology, branch lengths and missing data parameters. We derive a closed-form solution for the M-step in the case of no heritable missing data, and a block coordinate ascent approach in the general case which is more efficient than the standard General Time Reversible (GTR) phylogenetic model. On simulated data, LAML infers more accurate tree topologies and branch lengths than existing methods, with greater advantages on datasets with higher ratios of heritable to non-heritable missing data. We show that LAML provides unbiased time-scaled estimates of branch lengths. In contrast, we demonstrate that maximum parsimony methods for lineage tracing data not only underestimate branch lengths, but also yield branch lengths which are not proportional to time, due to the nonlinear decay in the number of mutations on branches further from the root. On lineage tracing data from a mouse model of lung adenocarcinoma, we show that LAML infers phylogenetic distances that are more concordant with gene expression data compared to distances derived from maximum parsimony. The LAML tree topology is more plausible than existing published trees, with fewer total cell migrations between distant metastases and fewer reseeding events where cells migrate back to the primary tumor. Crucially, we identify three distinct time epochs of metastasis progression, which includes a burst of metastasis events to various anatomical sites during a single month. Softwarehttps://github.com/raphael-group/LAML Data availabilityhttps://github.com/raphael-group/laml-experiments

systems biology↗

bioRxiv2022

Exploring Trade-offs in Scalable Phylogenetic Placement Methods

Phylogenetic placement is the problem of placing "query" sequences into an existing tree (called a "backbone tree"), and is useful in both microbiome analysis and to update large evolutionary trees. The most accurate phylogenetic placement method to date is the maximum likelihood-based method pplacer, which uses RAxML to estimate numeric parameters on the backbone tree and then adds the given query sequence to the edge that maximizes the probability that the resulting tree generates the query sequence. Unfortunately, pplacer fails to return valid outputs on many moderately large datasets, and so is limited to backbone trees with at most [~]10,000 leaves. In TCBB 2022, Wedell et al. introduced SCAMPP, a technique to enable pplacer to run on larger backbone trees. SCAMPP operates by finding a small "placement subtree" specific to each query sequence, within which the query sequence are placed using pplacer. That approach matched the scalability and accuracy of APPLES-2, the previous most scalable method. In this study, we explore a different aspect of pplacers strategy: the technique used to estimate numeric parameters on the backbone tree. We confirm anecdotal evidence that using FastTree instead of RAxML to estimate numeric parameters on the backbone tree enables pplacer to scale to much larger backbone trees, almost (but not quite) matching the scalability of APPLES-2 and pplacer-SCAMPP. We then evaluate the combination of these two techniques - SCAMPP and the use of FastTree. We show that this combined approach, pplacer-SCAMPP-FastTree, has the same scalability as APPLES-2, improves on the scalability of pplacer-FastTree, and achieves better accuracy than the comparably scalable methods. Availability: https://github.com/gillichu/PLUSplacer-taxtastic.

bioinformatics↗

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↗

bioRxiv2022

User-guided local and global copy-number segmentation for tumor sequencing data

Copy-number aberrations (CNAs) are genetic alterations that amplify or delete the number of copies of large genomic segments. Although they are ubiquitous in cancer and, thus, a critical area of current cancer research, CNA identification from DNA sequencing data is challenging because it requires partitioning of the genome into complex segments with the same copy-number states that may not be contiguous. Existing segmentation algorithms address these challenges either by leveraging the local information among neighboring genomic regions, or by globally grouping genomic regions that are affected by similar CNAs across the entire genome. However, both approaches have limitations: overclustering in the case of local segmentation, or the omission of clusters corresponding to focal CNAs in the case of global segmentation. Importantly, inaccurate segmentation will lead to inaccurate identification of important CNAs. For this reason, most pan-cancer research studies rely on manual procedures of quality control and anomaly correction. To improve copy-number segmentation and their control, we introduce CNAVO_SCPLOWIZC_SCPLOW, a web-based tool that enables the user to simultaneously perform local and global segmentation, thus overcoming the limitations of each approach. Using simulated data, we demonstrate that by several metrics, CNAVO_SCPLOWIZC_SCPLOW allows the user to obtain more accurate segmentation relative to existing local and global segmentation methods. Moreover, we analyze six bulk DNA sequencing samples from three breast cancer patients. By validating with parallel singlecell DNA sequencing data from the same samples, we show that by using CNAVO_SCPLOWIZC_SCPLOW, our user was able to obtain more accurate segmentation and improved accuracy in downstream copy-number calling. CNAVO_SCPLOWIZC_SCPLOW is available at https://github.com/elkebir-group/cnaviz.

genomics↗

bioRxiv2020

Reconstructing the history of founder events using genome-wide patterns of allele sharing across individuals

Founder events play a critical role in shaping genetic diversity, impacting the fitness of a species and disease risk in humans. Yet our understanding of the prevalence and distribution of founder events in humans and other species remains incomplete, as most existing methods for characterizing founder events require large sample sizes or phased genomes. To learn about the frequency and evolutionary history of founder events, we introduce ASCEND (Allele Sharing Correlation for the Estimation of Non-equilibrium Demography), a flexible two-locus method to infer the age and strength of founder events. This method uses the correlation in allele sharing across the genome between pairs of individuals to recover signatures of past bottlenecks. By performing coalescent simulations, we show that ASCEND can reliably estimate the parameters of founder events under a range of demographic scenarios, with genotype or sequence data. We apply ASCEND to ~5,000 worldwide human samples (~3,500 present-day and ~1,500 ancient individuals), and ~1,000 domesticated dog samples. In both species, we find pervasive evidence of founder events in the recent past. In humans, over half of the populations surveyed in our study had evidence for a founder events in the past 10,000 years, associated with geographic isolation, modes of sustenance, and historical invasions and epidemics. We document that island populations have historically maintained lower population sizes than continental groups, ancient hunter-gatherers had stronger founder events than Neolithic Farmers or Steppe Pastoralists, and periods of epidemics such as smallpox were accompanied by major population crashes. Many present-day groups--including Central & South Americans, Oceanians and South Asians--have experienced founder events stronger than estimated in Ashkenazi Jews who have high rates of recessive diseases due to their history of founder events. In dogs, we uncovered extreme founder events in most groups, more than ten times stronger than the median strength of founder events in humans. These founder events occurred during the last 25 generations and are likely related to the establishment of dog breeds during Victorian times. Our results highlight a widespread history of founder events in humans and dogs, and provide insights about the demographic and cultural processes underlying these events.

evolutionary biology↗