Biology subjects

Speed, T. P.

Publications and source records attributed to Speed, T. P..

At least 19 recordsLinked to original sources

bioRxiv2026

Strain resolution and wide relative abundance quantification of bacterial microbiomes with nanopore sequencing of the ribosomal RNA operon

Identification and quantification of microbiomes at strain level taxonomic resolution is increasingly sought. Longread sequencing of sufficiently high accuracy can provide this using ribosomal RNA operon (rrn) sequencing, subject to strains differing by at least one base in some of their operons. Prior studies of nanopore-sequenced amplicon reads have not established reliable strain level resolution and sensitive quantification. Here we demonstrate that if a sufficient number of rrn operon reads with a mean error rate of better than about 0.008 errors per base are available in a dataset, then strain identification and quantification, mediated by operon-level resolution, can often be achieved. Two multiple dataset sources of real, nanopore-sequenced bacterial mock microbiome were accessed, and multiple datasets of a realistic mock human gut microbiome, using state-of-the-art simulated nanopore-sequenced reads, were simulated. These datasets had read quality distributions lying between those of the real datasets, did not have multiple strains for any species, but, for the higher quality datasets, allowed demonstration of within-strain operon resolution and quantification over 2 orders of magnitude. The simulation-generated datasets demonstrated multiple cases of within-species strain resolution with accurate relative abundance quantification. The analysis of such datasets has been implemented here in a pipeline of 4 multi-parameter command line calls. This allows intra-species resolution and quantification over at least 3 orders of magnitude with more than 50 strains present in the microbiome

bioinformatics↗

bioRxiv2026

Fast, accurate and scalable normalization for RNA sequencing data with the RUVprps software package

Unwanted variation refers to any source of variability in the data that can compromise down-stream analysis. Effective removal of such variation from gene expression data is essential to derive accurate and meaningful biological results. We refer to this process as normalization. Data may come from a single study or from multiple studies with different sources of unwanted variation. We have previously developed the RUV-III method for normalizing omics data with a strong focus on transcriptomics. Initially, we introduced RUV-III for the normalization of Nanostring nCounter gene expression data, utilizing genuine technical replicates and pseudo-replicates as control samples. Subsequently, we proposed RUV-III with pseudo-replicates of pseudo-samples (PRPS), and which demonstrated its potential in mitigating the effects of different sources of unwanted variation in large and complex RNA-seq studies. To enhance accessibility and performance of this method, we present a new comprehensive R package named RUVprps. The package offers over 100 functions including ones for assessing variation in both biological and unwanted variables, an automated RUV-III normalization process, and metrics for evaluating the effectiveness of the resulting normalizations. Further, it introduces several new features such as ways of identifying unknown sources of unwanted variation, strategies to identify suitable negative control genes, and methods for generating PRPS when information on the biological and unwanted variation is unavailable. The package also implements a faster approach to RUV-III normalization, streamlining its application to large RNA-seq datasets. Our freely available R package and normalization assessment pipeline can help find effective data normalization methods for new data and help benchmark new methods.

bioinformatics↗

bioRxiv2025

CMDdemux: an efficient single cell demultiplexing method

Multiplexing technologies label cells with molecular tags, allowing cells from different donors to be pooled together for sequencing. Although this approach enhances cell throughput, eliminates batch effects, and enables doublet detection, limitations of hashtag-based labelling can still lead to low-quality data. Existing demultiplexing methods can accurately assign donor identities in high-quality datasets, but they often fail on low-quality data. To address this, we developed CMDdemux, a method comprising three key steps: within-cell centered log-ratio (CLR) normalization of hashtag count data, K-medoids clustering, and classification of cells based on Mahalanobis distance. By integrating both hashing and mRNA data, CMDdemux achieves high accuracy in distinguishing singlets, doublets, and negatives. It also provides visualization tools to help users inspect potentially misclassified droplets. We benchmarked CMDdemux against existing methods using a range of high- and low-quality datasets. Results show that CMDdemux consistently outperforms other approaches, demonstrating robust performance on both high- and low-quality data where other methods fail. CMDdemux is particularly effective in handling diverse types of low-quality multiplexing data across different multiplexing technologies.

bioinformatics↗

bioRxiv2025

Antigen reactivity defines tissue-resident memory and exhausted T cells in tumours

CD8+ T cells are a key weapon in the therapeutic armamentarium against cancer. While CD8+CD103+ T cells with a tissue-resident memory T (TRM) cell phenotype have been favourably correlated with patient prognoses1-6, the tumour microenvironment also contains dysfunctional exhausted T (TEX) cells that exhibit a myriad of TRM-like features, leading to conflation of these two populations. Here, we deconvolute TRM and TEX cells within the intratumoural CD8+CD103+ T cell pool across human cancers, ascribing markers and gene signatures that distinguish these CD8+ populations and enable their functional distinction. We found that while TRM cells exhibit superior functionality and are associated with long-term survival post-tumour resection, they are not associated with responsiveness to immune checkpoint blockade. Deconvolution of the two populations showed that tumour-associated TEX and TRM cells are clonally distinct, with the latter comprising both tumour-independent bystanders and tumour-specific cells segregated from their cognate antigen. Intratumoural TRM cells can be forced towards an exhausted fate when chronic antigen stimulation occurs, arguing that the presence or absence of continuous antigen exposure within the microenvironment is the key distinction between respective tumour-associated TEX and TRM populations. These results suggest unique roles for TRM and TEX cells in tumour control, underscoring the need for distinct strategies to harness these T cell populations in novel cancer therapies.

immunology↗

bioRxiv2024

Myeloid-instructed CD14+CD4+ T cells within the tumor microenvironment are associated with TNFα signaling and prolonged survival in non-small cell lung cancer

The tumor microenvironment (TME) harbors a diverse array of innate and adaptive immune cells that can either support anti-tumor immunity or facilitate tumor progression. Among these, myeloid cells are key drivers of immunosuppression, yet therapeutic strategies targeting their function have achieved limited clinical success. A deeper understanding of the mechanisms by which immunosuppressive myeloid cells promote tumor progression is needed to guide the development of more effective treatments. Using spatial multi-omics analyses, we identified a population of myeloid-instructed CD14+ T cells present in both the tumor core and adjacent non-malignant lung tissue of patients with non-small cell lung cancer (NSCLC). CD4+ T cells acquired CD14 from myeloid cells by trogocytosis, resulting in an atypical T cell phenotype. High infiltration of CD14+CD4+ T cells in the tumor core was associated with poor patient survival. Spatial transcriptomics profiling revealed that tumors enriched in CD14+CD4+ T cells exhibited increased TNF signaling. Functional assays demonstrated that TNF enhanced trogocytosis, promoting the transfer of myeloid membrane components to CD4{square} T cells and driving the accumulation of CD14{square}CD4{square} T cells. These findings uncover a novel TNF-mediated mechanism of immunosuppression in the TME, whereby TNF promotes aberrant myeloid-T cell interactions that may contribute to tumor progression. This work highlights a previously unrecognized axis of innate-adaptive immune crosstalk in NSCLC and suggests that targeting TNF could disrupt this pathway, restore effective T cell function, and improve therapeutic outcomes.

cancer biology↗

bioRxiv2024

GLMsim: a GLM-based single cell RNA-seq simulator incorporating batch and biological effects

With development of the single cell RNA-seq technologies, large numbers of cells can now be routinely sequenced by different platforms. This requires us to choose an efficient integration tool to merge those cells, and computational simulators to help benchmark and assess the performance of these tools. Although existing single cell RNA-seq simulators can simulate library size, biological and batch effects separately, they currently do not capture associations among these three factors. Here we present GLMsim, the first single cell RNA-seq simulator to simultaneously capture the library size, biology and unwanted variation and their associations via a generalized linear model, and to simulate data resembling the original experimental data in these respects. GLMsim is capable of quantitatively benchmarking different single cell integration methods, and assessing their abilities to retain biology and remove library size and batch effects.

bioinformatics↗

bioRxiv2024

Early cytokine-driven adaptation of survival pathways in lymphoid cells during targeted therapies

Venetoclax, a first-in-class BH3 mimetic drug targeting BCL-2, has improved outcomes for patients with chronic lymphocytic leukemia (CLL). Early measurements of the depth of the venetoclax treatment response, assessed by minimal residual disease, are strong predictors of long-term clinical outcomes. Yet, there are limited data concerning the early changes induced by venetoclax treatment that might inform strategies to improve responses. To address this gap, we conducted longitudinal mass cytometric profiling of blood cells from patients with CLL during the first two months of venetoclax monotherapy. At baseline, we resolved CLL heterogeneity at the single-cell level to define multiple subpopulations in all patients distinguished by proliferative, metabolic and cell survival proteins. Venetoclax induced significant reduction in all CLL subpopulations coincident with rapid upregulation of pro-survival BCL-2, BCL-XL and MCL-1 proteins in surviving cells, which had reduced sensitivity to the drug. Mouse models recapitulated the venetoclax-induced elevation of survival proteins in B cells and CLL-like cells that persisted in vivo, with genetic models demonstrating that extensive apoptosis and access to the B cell cytokine, BAFF, were essential. Accordingly, analysis of patients with CLL that were treated with a different targeted therapy, the anti-CD20 antibody obinutuzumab, also exhibited marked elevation of BAFF and increased pro-survival proteins in leukemic cells that persisted. Overall, these data highlight the rapid adaptation of CLL cells to targeted therapies via homeostatic factors and support co-targeting of cytokine signals to achieve deeper and more durable long-term responses. Key pointsO_LILeukaemic cells rapidly adapt to targeted therapy by elevating pro-survival protein expression. C_LIO_LICell attrition and increased bioavailability of homeostatic cytokines drive this heightened survival, highlighting avenues for more potent combination therapies. C_LI

cancer biology↗

bioRxiv2024

MAssively-Parallel Flow cytometry Xplorer (MAPFX): A Toolbox for Analysing Data from the Massively-Parallel Cytometry Experiments

Massively-Parallel Cytometry (MPC) experiments allow cost-effective quantification of more than 200 surface proteins at single-cell resolution. The Infinity Flow (Inflow) analysis protocol was developed to measure highly informative protein backbone markers on all cells in all wells distributed across three 96-well plates, along with well-specific exploratory protein infinity markers. Backbone markers can be used to impute the infinity markers on cells in all other wells using machine learning methods. This protocol offers unprecedented opportunities for more comprehensive classification of cell types. However, some aspects of the protocol can be improved, including methods for background correction and removal of unwanted variation. Here, we propose MAPFX as an end-to-end toolbox that carefully pre-processes the raw data from MPC experiments, and further imputes the missing infinity markers in the wells without those measurements. Our pipeline starts by performing background correction on raw intensities to remove the noise from electronic baseline restoration and fluorescence compensation by adapting a normal-exponential convolution model. Unwanted technical variation, from sources such as well effects, is then removed using a log-normal model with plate, column, and row factors, after which infinity markers are imputed using the informative backbone markers as predictors. The completed dataset can then be used for clustering and other statistical analyses. Unique features of our approach include performing background correction prior to imputation and removing unwanted variation from the data at the cell-level, while explicitly accounting for the potential association between biology and unwanted factors. We benchmark our pipeline against alternative pipelines and demonstrate that our approach is better at preserving biological signals, removing unwanted variation, and imputing unmeasured infinity markers.

bioinformatics↗

bioRxiv2024

Full 16S and 23S rRNA gene-based,strain-level resolution of the microbiota of a mock bacterial microbiome using ONT nanopore sequencing

PurposeThe feasibility of achieving near-strain taxonomic resolution and reliable microbiota strain-level abundance estimation using amplicon-based nanopore sequencing was investigated. MethodsDenoising was applied to separate 16S and 23S genes extracted from a metagenomic dataset generated by Sereika et al. [9] using nanopore sequencing. Denoising used Kumar et al.s [8] Robust Amplicon Denoising (RAD) and generated amplicon sequence variants (ASVs). The Sereika et al. data set was generated from the Zymo [10] D6322 7-bacterial species mock microbiome. A sub-sampling procedure generated additional datasets that allowed sensitivity assessment over multiple orders of relative abundance. Alignment to bespoke 16S and 23S rRNA gene databases, provided both identification and abundance information data for sub-species analyses. ResultsSub-species identification was clearly achieved, both with the "even" D6322 dataset, and with the 4 sub-sampled datasets in which 3-orders of magnitude relative abundances were present. Multiple ASVs of length approximately 2500 bases were generated which gave perfect alignments to reference 23S rRNA genes. Similarly for the approximately 1500 base long 16S rRNA genes. Despite strain ambiguity for some species, the strain of each species known to be present was identified in all but 1 (of 70) cases examined for all species. A process to merge the 16S and 23S rRNA genes results reduced ambiguity, allowing better sub-species resolution, and better species, and sub-species, relative cellular abundance estimates. ConclusionPrincipled methods for dealing with strain-level ambiguity in microbiota analysis have been developed that are generally applicable. Their application to nanopore sequencing with the current state of Oxford Nanopore Technology, together with state-of-the-art denoising, allows near strain resolution of bacterial microbiota identity, and improved estimates of species, and sub-species, relative abundances.

bioinformatics↗

bioRxiv2023

The current landscape and emerging challenges of benchmarking single-cell methods

With the rapid development of computational methods for single-cell sequencing data, benchmarking serves as a valuable resource. As the number of benchmarking studies surges, it is timely to assess the current state of the field. We conducted a systematic literature search and assessed 282 papers, including all 130 benchmark-only papers from the search and an additional 152 method development papers containing benchmarking. This collective effort provides the most comprehensive quantitative summary of the current landscape of single-cell benchmarking studies. We examine performances across nine broad categories, including often ignored aspects such as role of datasets, robustness of methods and downstream evaluation. Our analysis highlights challenges such as how to effectively combine knowledge across multiple benchmarking studies and in what ways can the community recognise the risk and prevent benchmarking fatigue. This paper highlights the importance of adopting a community-led research paradigm to tackle these challenges and establish best practice standards.

bioinformatics↗

bioRxiv2022

An open protocol for modeling T Cell Clonotype repertoires using TCRβ CDR3 sequences

T cell receptor (TCR) repertoires can be profiled using next generation sequencing (NGS) to monitor dynamical changes in response to disease and other perturbations. Several strategies for profiling TCRs have been recently developed with different benefits and drawbacks. Genomic DNA-based bulk sequencing, however, remains the most cost-effective method to profile TCRs. The major disadvantage of this method is the need for multiplex target amplification with a large set of primer pairs with potentially very different amplification efficiencies. One approach addressing this problem is by iteratively adjusting the concentrations of the primers based on their efficiencies, and then computationally correcting any remaining bias. Yet there are no standard, publicly available protocols to process and analyze raw sequencing data generated by this method. Here, we utilize an equimolar primer mixture and propose a single statistical normalization step that efficiently corrects for amplification bias post sequencing. Using samples analyzed by both approaches, we show that the concordance between bulk clonality metrics obtained from using the commercial kits and that developed herein is high. Therefore, we suggest the method presented here as an inexpensive and non-commercial alternative for measuring and monitoring adaptive dynamics in TCR clonotype repertoire.

bioinformatics↗

bioRxiv2022

Robust differential composition and variability analysis for multisample cell omics

Cell omics such as single-cell genomics, proteomics and microbiomics allow the characterisation of tissue and microbial community composition, which can be compared between conditions to identify biological drivers. This strategy has been critical to unveiling markers of disease progression such as cancer and pathogen infection. For cell omic data, no method for differential variability analysis exists, and methods for differential composition analysis only take a few fundamental data properties into account. Here we introduce sccomp, a generalised method for differential composition and variability analyses able to jointly model data count distribution, compositionality, group-specific variability and proportion mean-variability association, with awareness against outliers. Sccomp is an extensive analysis framework that allows realistic data simulation and cross-study knowledge transfer. Here, we demonstrate that mean-variability association is ubiquitous across technologies showing the inadequacy of the very popular Dirichlet-multinomial modelling and provide mandatory principles for differential variability analysis. We show that sccomp accurately fits experimental data, with a 50% incremental improvement over state-of-the-art algorithms. Using sccomp, we identified novel differential constraints and composition in the microenvironment of primary breast cancer. Significance statementDetermining the composition of cell populations is made possible by technologies like single-cell transcriptomics, CyTOF and microbiome sequencing. Such analyses are now widespread across fields (~800 publications/month, Scopus). However, existing methods for differential abundance do not model all data features, and cell-type/taxa specific differential variability is not yet possible. Increase in the variability of tissue composition and microbial communities is a well-known indicator of loss of homeostasis and disease. A suitable statistical method would enable new types of analyses to identify component-specific loss of homeostasis for the first time. This and other innovations are now possible through our discovery of the mean-variability association for compositional data. Based on this fundamental observation, we have developed a new statistical model, sccomp, that enables differential variability analysis for composition data, improved differential abundance analyses, with cross-sample information borrowing, outlier identification and exclusion, realistic data simulation, based on experimental datasets, cross-study knowledge transfer.

bioinformatics↗

bioRxiv2021

RUV-III-NB: Normalization of single cell RNA-seq Data

Despite numerous methodological advances, the normalization of single cell RNA-seq (scRNA-seq) data remains a challenging task and the performance of different methods can vary greatly across datasets. Part of the reason for this is the different kinds of unwanted variation, including library size, batch and cell cycle effects, and the association of these with the biology embodied in the cells. A normalization method that does not explicitly take into account cell biology risks removing some of the signal of interest. Furthermore, most normalization methods remove the effects of unwanted variation for the cell embedding used for clustering-based analysis but not from gene-level data typically used for differential expression (DE) analysis to identify marker genes. Here we propose RUV-III-NB, a statistical method that can be used to remove unwanted variation from both the cell embedding and gene-level counts. RUV-III-NB explicitly takes into account its potential association with biology when removing unwanted variation via the use of pseudo-replicates. The method can be used for both UMI or sequence read counts and returns adjusted counts that can be used for downstream analyses such as clustering, DE and pseudotime analyses. Using five publicly available datasets that encompass different technological platforms, kinds of biology and levels of association between biology and unwanted variation, we show that RUV-III-NB manages to remove library size and batch effects, strengthen biological signals, improve differential expression analyses, and lead to results exhibiting greater concordance with independent datasets of the same kind. The performance of RUV-III-NB is consistent across the five datasets and is not sensitive to the number of factors assumed to contribute to the unwanted variation. It also shows promise for removing other kinds of unwanted variation such as platform effects. The method is implemented as a publicly available R package available from https://github.com/limfuxing/ruvIIInb.

bioinformatics↗

bioRxiv2021

Removing unwanted variation from large-scale cancer RNA-sequencing data

The accurate identification and effective removal of unwanted variation are essential to derive meaningful biological results from RNA-seq data, especially when the data come from large and complex studies. We have used The Cancer Genome Atlas (TCGA) RNA-seq data to show that library size, batch effects, and tumor purity are major sources of unwanted variation across all TCGA RNA-seq datasets and that existing gold standard approaches to normalizations fail to remove this unwanted variation. Additionally, we illustrate how different sources of unwanted variation can compromise downstream analyses, including gene co-expression, association between gene expression and survival outcomes, and cancer subtype identifications. Here, we propose the use of a novel strategy, pseudo-replicates of pseudo-samples (PRPS), to deploy the Removing Unwanted Variation III (RUV-III) method to remove different sources of unwanted variation from large and complex gene expression studies. Our approach requires at least one roughly known biologically homogenous subclass of samples shared across sources of unwanted variation. To create PRPS, we first need to identify the sources of unwanted variation, which we will call batches in the data. Then the gene expression measurements of biologically homogeneous sets of samples are averaged within batches, and the results called pseudo-samples. Pseudo-samples with the same biology and different batches are then defined to be pseudo-replicates and used in RUV-III as replicates. The variation between pseudo-samples of a set pseudo-replicates is mainly unwanted variation. We illustrate the value of our approach by comparing it to the TCGA normalizations on several TCGA RNA-seq datasets. RUV-III with PRPS can be used for any large genomics project involving multiple labs, technicians, or platforms.

bioinformatics↗

bioRxiv2021

Early immune pressure imposed by tissue resident memory T cells sculpts tumour evolution in non-small cell lung cancer

Tissue-resident memory T cells (TRM) provide immune defence against local infection and can inhibit cancer progression. However, it is unclear to what extent chronic inflammation impacts TRM activation and how the immune pressure exerted by TRM affects developing tumours in humans. We performed deep profiling of lung cancers arising in never-smokers (NS) and ever-smokers (ES), finding evidence of enhanced TRM immunosurveillance in ES lung. Only tumours arising in ES patients underwent clonal immune escape, even when evaluating cancers with similar tumour mutational burden to NS patients, suggesting that the timing of immune pressure exerted by TRM is a critical factor in the evolution of tumour immune evasion. Tumours grown in T cell quiescent NS lungs displayed little evidence of immune evasion and had fewer neoantigens with low diversity, paradoxically making them amenable to treatment with agonist of the costimulatory molecule, ICOS. These data demonstrate local environmental insults enhance TRM immunosurveillance of human tissue, shape the evolution of tumour immunogenicity and that this interplay informs effective immunotherapeutic modalities.

cancer biology↗

bioRxiv2020

hRUV: Hierarchical approach to removal of unwanted variation for large-scale metabolomics data

Liquid chromatography-mass spectrometry based metabolomics studies are increasingly applied to large population cohorts, running for several weeks to months, even extending to years of data acquisition. This inevitably introduces unwanted intra- and inter-batch variations over time that can overshadow true biological signals and thus hinder potential biological discoveries. To date, normalization approaches have struggled to mitigate the variability introduced by technical factors whilst preserving biological variance, especially for protracted acquisitions. Here, we designed an experiment with an arrangement to embed biological sample replicates to measure the variance within and between batches for over 1,000 human plasma samples run over 44 days. We integrate these replicates in a novel workflow to remove unwanted variation in a hierarchical structure (hRUV) by progressively merging the adjustments in neighbouring batches. We demonstrate significant improvement of hRUV over existing methods in maintaining biological signals whilst removing unwanted variation for large scale metabolomics studies.

bioinformatics↗

bioRxiv2020

The ratio of exhausted to resident infiltrating lymphocytes is prognostic for colorectal cancer patient outcome

Immunotherapy success in colorectal cancer (CRC) is mainly limited to patients whose tumours exhibit high microsatellite instability (MSI). However, there is variability in treatment outcomes within this group, which is in part driven by the frequency and characteristics of tumour infiltrating immune cells. Indeed, the presence of specific infiltrating immune cell subsets has been shown to correlate with immunotherapy responses and is in many cases prognostic of treatment outcome. Tumour-infiltrating lymphocytes (TILs) can undergo distinct differentiation programs, acquiring features of tissue-residency or exhaustion, a process during which T cells upregulate inhibitory receptors such as PD-1 and loose functionality. While residency and exhaustion programs of CD8 T cells are relatively well-studied, these programs have only recently been appreciated in CD4 T cells and remain largely unknown in tumour-infiltrating natural killer (NK) cells. In this study, we use single cell RNA-seq data to identify signatures of residency and exhaustion in CRC infiltrating lymphocytes, including CD8, CD4 and NK cells. We then test these signatures in independent single cell data from tumour and normal tissue infiltrating immune cells. Further, we use versions of these signatures designed for bulk RNA-seq data to explore tumour intrinsic mutations associated with residency and exhaustion from TCGA data. Finally, using two independent transcriptomic data sets from patients with colon adenocarcinoma, we show that combinations of these signatures, in particular combinations of NK activity signatures, together with tumour-associated signatures, such as TGF-{beta} signalling, are associated with distinct survival outcomes in colorectal cancer patients.

cancer biology↗

bioRxiv2020

Strain-level sample characterisation using long reads and MAPQ scores

AO_SCPLOWBSTRACTC_SCPLOWA simple but effective method for strain-level characterisation of microbial samples using long read data is presented. The method, which relies on having a non-redundant database of reference genomes, differentiates between strains within species and determines their relative abundance. It provides markedly better strain differentiation than that reported for the latest long read tools. Good estimates of relative abundances of highly similar strains present at less than 1% are achievable with as little as 1Gb of reads. Host contamination can be removed without great loss of sample characterisation performance. The method is simple and highly flexible, allowing it to be used for various different purposes, and as an extension of other characterisation tools. A code body implementing the underlying method is freely available.

bioinformatics↗