Dinithi Sumanaweera
A5019059813- Machine Learning in Bioinformatics
- Single-cell and spatial transcriptomics
- Genomics and Phylogenetic Studies
- Algorithms and Data Compression
- SARS-CoV-2 and COVID-19 Research
- RNA and protein synthesis mechanisms
- Protein Structure and Dynamics
- Gene Regulatory Network Analysis
- COVID-19 Clinical Research Studies
- T-cell and B-cell Immunology
- Gene expression and cancer classification
- Immunotherapy and Immune Responses
- Biomedical Ethics and Regulation
- Osteoarthritis Treatment and Mechanisms
- Bioinformatics and Genomic Networks
- Immune responses and vaccinations
- Cancer-related molecular mechanisms research
- Fractal and DNA sequence analysis
- Inflammatory Bowel Disease
- IL-33, ST2, and ILC Pathways
- Neurogenetic and Muscular Disorders Research
- Parkinson's Disease Mechanisms and Treatments
- Immune Cell Function and Interaction
- Parallel Computing and Optimization Techniques
- RNA Research and Splicing
University of Cambridge
2024
Wellcome Sanger Institute
2023-2024
Monash University
2018-2022
University of Moratuwa
2015-2019
The COVID-19 pandemic is an ongoing global health threat, yet our understanding of the dynamics early cellular responses to this disease remains limited
Human prenatal skin is populated by innate immune cells, including macrophages, but whether they act solely in immunity or have additional functions morphogenesis unclear. Here we assembled a comprehensive multi-omics reference atlas of human (7-17 post-conception weeks), combining single-cell and spatial transcriptomics data, to characterize the microanatomical tissue niches skin. This revealed that crosstalk between non-immune cells underpins formation hair follicles, implicated scarless...
Amyotrophic lateral sclerosis (ALS) is a fatal neurodegenerative disease where substantial heterogeneity in clinical presentation urgently requires better stratification of patients for the development drug trials and care. In this study we explored through crowdsourcing approach, DREAM Prize4Life ALS Stratification Challenge. Using data from >10,000 1479 community-based patient registers, more than 30 teams developed new approaches machine learning clustering, outperforming best current...
Abstract The COVID-19 pandemic is an ongoing global health threat, yet our understanding of the cellular disease dynamics remains limited. In unique human challenge study we used single cell genomics nasopharyngeal swabs and blood to temporally resolve abortive, transient sustained infections in 16 seronegative individuals challenged with preAlpha-SARS-CoV-2. Our analyses revealed rapid changes type proportions dozens highly dynamic response states epithelial immune cells associated specific...
Abstract The gastrointestinal tract is a multi-organ system crucial for efficient nutrient uptake and barrier immunity. Advances in genomics surge diseases 1,2 has fuelled efforts to catalogue cells constituting tissues health disease 3 . Here we present systematic integration of 25 single-cell RNA sequencing datasets spanning the entire healthy development adulthood. We uniformly processed 385 samples from 189 controls using newly developed automated quality control approach (scAutoQC),...
Alignments are correspondences between sequences. How reliable alignments of amino acid sequences proteins, and what inferences about protein relationships can be drawn? Using techniques not previously applied to these questions, by weighting every possible sequence alignment its posterior probability we derive a formal mathematical expectation, develop an efficient algorithm for computation the distance alternative allowing quantitative comparisons sequence-based with corresponding...
Abstract Single-cell data analysis can infer dynamic changes in cell populations, for example across time, space or response to perturbation. To compare these dynamics between two conditions, trajectory alignment via programming (DP) optimization is frequently used, but limited by assumptions such as a definite existence of match. Here we describe Genes2Genes , Bayesian information-theoretic DP framework aligning single-cell trajectories. overcomes current limitations and able capture...
Abstract The information criterion of minimum message length (MML) provides a powerful statistical framework for inductive reasoning from observed data. We apply MML to the problem protein sequence comparison using finite state models with Dirichlet distributions. resulting allows us supersede ad hoc cost functions commonly used in field, by systematically addressing arbitrariness alignment parameters, and disconnect between substitution scores gap costs. Furthermore, our enables generation...
Summary The consistent production of in vitro chondrocytes that faithfully recapitulate vivo development would be great benefit for musculoskeletal disease modelling and regenerative medicine. Current efforts are often limited by off-target differentiation, resulting a heterogeneous product. Furthermore, the lack comparison to human embryonic tissue, precludes detailed evaluation cells. Here, we perform single-cell RNA sequencing long bones dissected from first trimester hind limbs range...
Comparison of protein sequences via alignment is an important routine in modern biological studies. Although the technologies for aligning proteins are mature, current state art continues to be plagued by many shortcomings, chiefly due reliance on: (i) naive objective functions, (ii) fixed substitution scores independent being considered, (iii) arbitrary choices gap costs, and (iv) reporting, often, one optimal without a way recognise other competing sequence alignments. Here, we address...
Abstract Introduction and aims Molecular regulation of mammalian skin development has been largely derived from murine studies. The lack human studies stems primarily the difficulties in prenatal access. There is an important need to study owing notable differences between mouse paucity mechanistic understanding congenital disorders. Methods We assembled first comprehensive multiomic reference atlas (21 samples 7 16 postconception weeks), combining single-cell (235 201 cells) spatial...
Summary Human prenatal skin is populated by innate immune cells including macrophages, and whether they act solely in immunity or have additional functions morphogenesis unclear. We assembled the first comprehensive multi-omic reference atlas of human (7-16 post-conception weeks), combining single cell spatial transcriptomic data, to characterise skin’s microenvironmental cellular organisation. This revealed that crosstalk between non-immune underpins formation hair follicles, has...
Abstract Summary Sequences of proteins evolve by accumulating substitutions together with insertions and deletions (indels) amino acids. However, it remains a common practice to disconnect indels, infer approximate models for each them separately, quantify sequence relationships. Although this approach brings computational convenience (which its primary motivation), there is dearth attempts unify model systematically together. To overcome gap, article demonstrates how complete statistical...
Many computer systems; especially in corporations, contain large amount of documents such as letters, reports and presentations. are present several versions. Such data needs to be synchronized with branch offices mobile devices, often over slow expensive connections. However, many stored an already compressed format, it is difficult compress them further by exploiting the hidden redundancies. We a novel approach named RepoZip which improves compression existing algorithm document...
This work demonstrates how a complete statistical model quantifying the evolution of pairs aligned proteins can be constructed from time-parameterised substitution matrix and 3-state alignment machine. All parameters such inferred any benchmark data-set protein sequences. allows us to examine nine well-known matrices on six benchmarks curated using various structural methods; that does not explicitly "time"-dependent Markov process is converted corresponding base-matrix does. In addition,...
Single-cell RNA-seq datasets are growing in size and complexity, enabling the study of cellular composition changes various biological/clinical contexts. Scalable dimensionality reduction techniques need to disentangle biological variation them, while accounting for technical confounders. In this work, we extend a popular approach probabilistic non-linear reduction, Gaussian process latent variable model, scale massive single-cell explicitly The key idea is use an augmented kernel which...
Protein function annotation is vital for identifying disease causative factors and solving mysteries behind biological system complexities. As manual requires costly laborious in vitro methods, silico protein prediction preferred nowadays. According to literature, one five yeast mitochondrial proteins are known be human related. This paper presents a genetic algorithmically weighted heterogeneous data ensemble classify Saccharomyces cerevisiae under 'mitochondrion organization'(GO:0007005)...
This paper presents a framework to characterize and identify local sequences of proteins that are statistically redundant under the measure Shannon information content while accounting for variations in their occurrences over evolutionary insertions, deletions, substitutions amino acids. The identification such provides insights downstream studies on proteins. Here, we have applied our methods acid sequence data sets derived from database corresponding 935,552 substructural regions varying...