Biology subjects

Mallon, A.-M.

Publications and source records attributed to Mallon, A.-M..

10 recordsLinked to original sources

bioRxiv2021

A convolutional neural network for the quality control of MRI defacing

Large-scale neuroimaging datasets present unique challenges for automated processing pipelines. Mo-tivated by a large clinical trials dataset of Multiple Sclerosis with over 235,000 MRI scans, we consider the challenge of defacing - anonymisation to remove identifying facial features. The defacing process must undergo quality control (QC) checks to ensure that the facial features have been adequately anonymised and that the brain tissue is left completely intact. Visual QC checks - particularly on a project of this scale - are time-consuming and can cause delays in preparing data. We have developed a convolutional neural network (CNN) that can assist with the QC of the application of MRI defacing; our CNN is able to distinguish between scans that are correctly defaced, and can also classify defacing failures into three sub-types to facilitate parameter tuning during remedial re-defacing. Since integrating the CNN into our anonymisation pipeline, over 75,000 scans have been processed. Strict thresholds have been applied so that ambiguous classifications are referred for visual QC checks, however all scans still undergo a time-efficient verification check before being marked as passed. After applying the thresholds, our network is 92% accurate and is able to classify nearly half of the scans without the need for protracted manual checks. Even with the introduction of the verification checks, incorporation of the CNN has reduced the time spent undertaking QC checks by 42% during initial defacing, and by 35% overall. With the help of the CNN, we have been able to successfully deface 96% of the scans in the project whilst maintaining high QC standards. In a similarly sized new project of this scale, we would expect the model to reduce the time spent on manual QC checks by 125 hours. Our approach is applicable to other projects with the potential to greatly improve the efficiency of imaging anonymisation pipelines.

neuroscience↗

bioRxiv2021

Mechanisms of DNA repair have evolved to minimise the probability of nonsense mutations.

Variation in sequence mutability has important implications for evolutionary models and predicting disease occurrence, and is driven in part by evolutionary divergence in mechanisms of DNA repair. The aim of this study was twofold: first, to assess the effect of local sequence context on substitution rates in the mouse lineage; second, to investigate the relationship between sequence mutability and selection. We show that the 7-mer context (i.e three bases either side of the base of interest) explains more variation in substitution rates between chromosomes in the mouse lineage than either the 3-mer, 5-mer, or 9-mer contexts. Furthermore, we also show that 7-mer substitutions with the potential to cause nonsense mutations when they occur in translated sequences occur at a lower rate across the genome than 7-mer substitutions with the potential to cause synonymous mutations. We propose that mechanisms of DNA repair have evolved to prioritise substitutions that are more likely to be deleterious to fitness.

genomics↗

bioRxiv2021

Phenotyping of Klf14 mouse white adipose tissue enabled by whole slide segmentation with deep neural networks

White adipose tissue (WAT) plays a central role in metabolism, with multiple diseases and genetic mutations causing its remodeling. Quantitative analysis of white adipocyte size is of great interest to understand physiology and disease, but previous studies of H&E histology have been limited to a subsample of whole depot cross-sections. In this paper, we present the deep learning pipeline DeepCytometer, that can segment mouse and human whole slides ([~=]40,000 cells per mouse slide on average) using an adaptive tiling method, correct for cell overlap and reject non-white adipocytes from the segmentation. Using quantile colour maps we show intra- and inter-depot cell size heterogeneity with local correlation; quantile estimates also suggest significant differences in population estimates from 75 whole slides compared to smaller data sets. We propose three linked levels (body weight BW, depot weight DW and cell area quartiles) for exploratory analysis of mouse Klf14 phenotypes in gonadal and subcutaneous depots. We find a rich set of phenotypes when stratifying by sex, depot and three genotype strata: (1) WTs/Hets with a Het father (Controls), (2) WTs with a Het mother, and (3) Hets with a Het mother (functional KOs or FKOs). Namely, at BW level, mean difference testing suggests that female FKOs are similar to Controls, but WTs with a Het mother are significantly larger. At DW and cell levels, linear models with interaction terms and BW or DW covariates, respectively, reveal phenotypes not shown by difference of means tests. For example, at DW level, gonadal and subcutaneous models are similar, and female FKOs have lower fat percentage than Controls due to both an offset and the DW/BW slope in the linear model. Meanwhile, female WTs with a Het mother have on average similar fat percentage to Controls, but as their slopes are close to zero, their DWs are uncorrelated to BW, suggesting that larger female WTs with a Het mother have lower fat percentage than smaller ones. In contrast to depot level, at cell level female gonadal phenotypes diverge from subcutaneous ones. Furthermore, male Controls and FKOs have similar average area values in subcutaneous depots, but area~DW slope flattening in FKOs suggests that larger DWs could be caused by cell size increase in Controls and by cell count increase in FKOs. Thus, DeepCytometer and associated exploratory analysis reveal new insights into adipocyte heterogeneity and phenotyping.

bioinformatics↗

bioRxiv2021

Comprehensive phenotypic analysis of the Dp1Tyb mouse strain reveals a broad range of Down Syndrome-related phenotypes

Down syndrome (DS), trisomy 21, results in many complex phenotypes including cognitive deficits, heart defects and craniofacial alterations. Phenotypes arise from an extra copy of human chromosome 21 (Hsa21) genes. However, causative genes remain mostly unknown. Animal models enable identification of these genes and pathological mechanisms. The Dp1Tyb mouse model of DS has an extra copy of 63% of Hsa21-orthologous mouse genes. Here, we comprehensively phenotype Dp1Tyb mice and find wide-ranging DS-like phenotypes including aberrant megakaryopoiesis, reduced bone density, and deficits in memory, locomotion, hearing and sleep. Thus, Dp1Tyb mice are an excellent model for studies of many complex DS phenotypes.

genetics↗

bioRxiv2020

Making sense of the linear genome, gene function and TADs

BackgroundTopologically associating domains (TADs) are thought to act as functional units in the genome. TADs co-localise genes and their regulatory elements as well as forming the unit of genome switching between active and inactive compartments. This has led to the speculation that genes which are required for similar processes may fall within the same TADs, allowing them to share regulatory programs and efficiently switch between chromatin compartments. However, evidence to link genes within TADs to the same regulatory program is limited. ResultsWe investigated the functional similarity of genes which fall within the same TAD. To do this we developed a TAD randomisation algorithm to generate sets of "random TADs" to act as null distributions. We found that while pairs of paralogous genes are enriched in TADs overall, they are depleted in TADs with CCCTC-binding factor (CTCF) ChIP-seq peaks at both boundaries. By assessing gene constraint as a proxy for functional importance we found that genes which singly occupy a TAD have greater functional importance than genes which share a TAD, and these genes are enriched for developmental processes. We found little evidence that pairs of genes in CTCF bound TADs are more likely to be co-expressed or share functional annotations than can be explained by their linear proximity alone. ConclusionsThese results suggest that algorithmically defined TADs consist of two functionally different groups, those which are bound by CTCF and those which are not. We detected no association between genes sharing the same CTCF TADs and increased co- expression or functional similarity, other than that explained by linear genome proximity. We do however find that functionally important genes are more likely to fall within a TAD on their own suggesting that TADs play an important role in the insulation of these genes.

genomics↗

bioRxiv2020

OpenStats: A Robust and Scalable Software Package for Reproducible Analysis of High-Throughput Phenotypic Data

Reproducibility in the statistical analyses of data from high-throughput phenotyping screens requires a robust and reliable analysis foundation that allows modelling of different possible statistical scenarios. Regular challenges are scalability and extensibility of the analysis software. In this manuscript, we describe OpenStats, a freely available software package that addresses these challenges. We show the performance of the software in a high-throughput phenomic pipeline in the International Mouse Phenotyping Consortium (IMPC) and compare the agreement of the results with the most similar implementation in the literature. OpenStats has significant improvements in speed and scalability compared to existing software packages including a 13-fold improvement in computational time to the current production analysis pipeline in the IMPC. Reduced complexity also promotes FAIR data analysis by providing transparency and benefiting other groups in reproducing and re-usability of the statistical methods and results. OpenStats is freely available under a Creative Commons license at www.bioconductor.org/packages/OpenStats.

bioinformatics↗

bioRxiv2020

LAMA: Automated image analysis for developmentalphenotyping of mouse embryos

Advanced 3D imaging modalities such as micro computed tomography (micro-CT), high resolution episcopic microscopy (HREM), and optical projection tomography (OPT) have been readily incorporated into high-throughput phenotyping pipelines, such as the International Mouse Phenotyping Consortium (IMPC). Such modalities generate large volumes of raw data that cannot be immediately harnessed without significant resources of manpower and expertise. Thus, rapid automated analysis and annotation is critical to ensure that 3D imaging data is able to be integrated with other multi-dimensional phenotyping data. To this end, we present an automated computational mouse phenotyping pipeline called LAMA, based on image registration, which requires minimal technical expertise and human input to use. Designed predominantly for developmental biologists, our software performs image pre-processing, registration, statistical and gene function annotation, and segmentation of 3D micro-CT data. We address several limitations of current methods and create an easy to use, fast solution application for mouse embryo phenotyping. We also present a highly granular, novel anatomical E14.5 (14.5 days post coitus) atlas of a population average that integrates with our pipeline to allow a range of dysmorphologies to be automatically annotated as well as results from the validation of the pipeline.

bioinformatics↗

bioRxiv2019

A resource of targeted mutant mouse lines for 5,061 genes

The International Mouse Phenotyping Consortium reports the generation of new mouse mutant strains for over 5,000 genes from targeted embryonic stem cells on the C57BL/6N genetic background. This includes 2,850 null alleles for which no equivalent mutant mouse line exists, 2,987 novel conditional-ready alleles, and 4,433 novel reporter alleles. This nearly triples the number of genes with reporter alleles and almost doubles the number of conditional alleles available to the scientific community. When combined with more than 30 years of community effort, the total mutant allele mouse resource covers more than half of the genome. The extensively validated collection is archived and distributed through public repositories, facilitating availability to the worldwide biomedical research community, and expanding our understanding of gene function and human disease.

genetics↗

bioRxiv2019

Human and mouse essentiality screens as a resource for disease gene discovery

Although genomic sequencing has been transformative in the study of rare genetic diseases, identifying causal variants remains a considerable challenge that can be addressed in part by new gene-specific knowledge. Here, we integrate measures of how essential a gene is to supporting life, as inferred from the comprehensive viability and phenotyping screens performed on knockout mice by the International Mouse Phenotyping Consortium and from human cell line essentiality screens. We propose a novel, cross-species gene classification across the Full Spectrum of Intolerance to Loss-of-function (FUSIL) and demonstrate that genes in five mutually exclusive FUSIL categories have differing characteristics in the biological processes they regulate, tissue expression levels and human mutation rates. Most notably, Mendelian disease genes, particularly those associated with developmental disorders, are highly overrepresented in the developmental lethal category, representing genes not essential for cell survival but required for organism development. Exploiting this finding, we have screened developmental disorder cases from three independent disease sequencing consortia and identified potentially pathogenic, de novo variants shared in different patients for several developmental lethal genes that have not previously been associated with rare disease. We therefore propose FUSIL as an efficient resource for disease gene discovery.

genomics↗

bioRxiv2019

Tolerance of nonsynonymous variation is closely correlated between human and mouse orthologues

Genic constraint describes how tolerant a gene is of nonsynonymous variation before it is removed from the population by negative selection. Here, we provide the first estimates of intraspecific constraint for mouse genes genome-wide, and show constraint is positively correlated between human and mouse orthologues (r = 0.806). We assess the relationships between mouse gene constraint and knockout phenotypes, showing gene constraint is positively associated with pleiotropy (ie an increased number of phenotype associations (R2 = 0.65)), in addition to an enrichment in lethal, developmental, and craniofacial knockout phenotypes amongst the most constrained genes. Finally, we show mouse constraint can be used to predict human genes associated with Mendelian disease, and is positively correlated with an increase in the number of known pathogenic variants in the human orthologue (R2 = 0.23). Our metrics of mouse and human constraint are available to inform future research using mouse models.

genomics↗