Upcoming events and workshops

October 14 - Harnessing the power of AI in biotechnology, personalised medicine, drug development and beyond

Speaker: Prof David Ascher (University of Queensland)

Abstract: We have developed a comprehensive computational platform that uses graph-based signatures to represent the wild-type environment of a residue in order to predict the structural and functional effects of mutations. This platform has been used to explore the effects of genetic disease and drug resistance mutations on protein folding, stability, dynamics and interactions, and their links to mutational tolerance and phenotypes. Mutations leading to larger molecular consequences, tended to be rarer, and needed the presence of compensatory mutations balancing these fitness costs to become fixed in a population.

We have now successfully clinically translated methods that use insights on the 3D effects of mutations to guide patient risk management in genetic diseases, and in the pre-emptive detection of drug resistance mutations in tuberculosis (rifampicin and pyrazinamide resistance). It has also been applied as part of drug development pipelines to guide design of drugs less prone to resistance.

This work has highlighted that structural bioinformatics tools, when applied in a systematic, integrated way, can provide a powerful and scalable approach for predicting structural and functional consequences of mutations in order to reveal molecular mechanisms leading to clinical and experimental phenotypes. These computational tools are freely available (http://biosig.unimelb.edu.au/biosig/tools).

About the speaker: David Ascher is Deputy Associate Dean of Research and Director of Biotechnology Program at The University of Queensland, Deputy Director of the Australian Centre for Ecogenomics and Head of the Computational Biology and Clinical Informatics laboratory at the Baker Institute. David’s research focus is in modelling biological data to gain insight into fundamental biological processes. One of his primary research interests has been developing tools to unravel the link between genotype and phenotype, using computational and experimental approaches to understand the effects of mutations on protein structure and function. His group has developed a platform of 65 widely used programs for assessing the molecular consequences of coding variants. This platform has now been successfully translated into clinical use to guide the diagnosis, management and treatment of a number of hereditary diseases, rare cancers and drug resistant infections, and into industry as part of drug development pipelines to guide design of safer and more effective therapeutics.

September 30 - Chatting About Pathology

Speaker: Dr Drew Williamson (Emory University)

Abstract: Generative models such as ChatGPT have catapulted AI further into the mainstream than ever before, while interested parties in healthcare have begun to test these models in medical settings. However, these models are rarely purpose-built for medical applications and the majority of those that do exist are targeted at a broader segment of healthcare than any individual provider practices. I’ll discuss the design, training, and evaluation of PathChat, our multimodal generative model created specifically for pathology. I’ll also discuss possible future directions in this line of inquiry.

About the speaker: Drew Williamson completed a BA in Mathematics at Oberlin College, an MD at Case Western Reserve University, residency in Anatomic Pathology at Brigham & Women’s Hospital, and fellowships in Molecular Genetic Pathology and Clinical Informatics at Mass General Brigham. This training was followed by postdoctoral research in the lab of Faisal Mahmood, PhD at Brigham & Women’s Hospital and Harvard Medical School. A board certified pathologist, Drew's research focuses on the applications of deep learning-based methods to pathology data, from histology images to genomics to, most recently, natural language. He is now an Assistant Professor in the department of Pathology & Laboratory Medicine at Emory University.

September 23 - How to catch a virus (and write software that biologists actually use)

Speaker: Dr Laura Luebbert (Broad Institute)

Abstract: There are an estimated 300,000 mammalian viruses from which infectious diseases in humans may arise. They inhabit human tissues such as the lungs, blood, and brain and often remain undetected. Efficient and accurate detection of viral infection is vital to understanding its impact on human health and to make accurate predictions to limit adverse effects, such as future epidemics. The increasing use of high-throughput sequencing methods in research, agriculture, and healthcare provides an opportunity for the cost-effective surveillance of viral diversity and investigation of virus-disease correlation. However, existing methods for identifying viruses in sequencing data rely on and are limited to reference genomes or cannot retain single-cell resolution through cell barcode tracking. We introduce a method that accurately and rapidly detects viral sequences in bulk and single-cell transcriptomics data based on highly conserved amino acid domains, which enables the detection of RNA viruses covering up to 10^12 virus species. The analysis of viral presence and host gene expression in parallel at single-cell resolution allows for the characterization of host viromes and the identification of viral tropism and host responses. We applied our method to identify putative novel viruses in rhesus macaque PBMC data that display cell type specificity and whose presence correlates with altered host gene expression.

About the speaker: Laura recently completed her Ph.D. in Computational Biology at the California Institute of Technology in the US. She recently wrote a pipeline for the detection of previously unknown viral sequences in NGS data and will apply her work towards the detection of emerging pathogens as a postdoctoral researcher in the laboratory of Prof. Pardis Sabeti at the Broad Institute of MIT and Harvard (with a joint affiliation with Harvard University). Laura is passionate about bridging the gap between biology and bioinformatics, having worked in both wet lab and computer science roles. Her first software project, gget, which facilitates access to large genomic databases, has been downloaded over 100,000 times and has become a worldwide standard in transcriptomic and proteomic data processing.

September 16 - Cell-type-specific mapping of enhancer-gene pairs from multimodal single cell data

Speaker: Dr Chang Su (Emory University)

Abstract: Mapping enhancers and target genes in disease-related cell types has provided critical insights into the functional mechanisms of genetic variants identified by Genome-wide associated studies (GWAS). However, most existing analyses rely on bulk data, which may overlook cell-type-specific enhancers and target genes. Recently, single-cell multimodal data measuring both gene expression and chromatin accessibility within the same cells have enabled the inference of enhancer-gene pairs in a cell-type-specific and context-specific manner. However, this task is challenged by the data's high sparsity, sequencing depth variation, and the computational burden of analyzing a large number of enhancer-gene pairs. To address these challenges, we propose scMultiMap, a statistical method that infers enhancer-gene association from sparse multimodal counts using a joint latent-variable model. It adjusts for confounding, permits fast moment-based estimation and provides analytically derived p-values. In systematic analyses of blood and brain data, scMultiMap shows appropriate type I error control, high statistical power with greater reproducibility across independent datasets and stronger consistency with orthogonal data modalities. Meanwhile, its computational cost is less than 1% of existing methods. When applied to study Alzheimer's disease (AD), scMultiMap gave the highest heritability enrichment in microglia and revealed new insights into the regulatory mechanisms of AD GWAS variants in microglia.

About the speaker: Chang Su is an Assistant Professor in Biostatistics and Bioinformatics at Emory University. Her research aims to develop statistical methodologies to address interesting biology questions with single-cell genomics and genetics data. Her current research topics include gene regulation in specific contexts and the genetics basis of single-cell genomics.

September 2 - Statistical identification of cell type-specific spatially variable genes in spatial transcriptomics

Speaker: Dr Lulu Shang (MD Anderson Cancer Center)

Abstract: Spatial transcriptomics is a collection of genomic technologies that enable transcriptomic profiling of tissues with spatial localization information. An essential task in spatial transcriptomics involves identifying genes with spatial expression patterns, known as spatially variable genes (SVGs). Importantly, a subset of SVGs displays diverse spatial expression patterns within a given cell type, thus representing key transcriptomic signatures underlying cellular heterogeneity. Here, we present Celina, a statistical method for systematically detecting this subset of cell type-specific SVGs (ct-SVGs). Celina utilizes a spatially varying coefficient model to accurately capture each gene's spatial expression pattern in relation to the distribution of cell types across tissue locations, ensuring effective type I error control and high statistical power. We evaluated the performance of Celina through comprehensive simulations and applications to five real datasets, where we also adapted and examined existing methods originated from other analytic settings to detect ct-SVGs. Celina proves powerful compared to these ad hoc method adaptations in single cell resolution spatial transcriptomics and stands as the only effective solution for spot resolution spatial transcriptomics. The ct-SVGs detected by Celina also enable novel biologically informed downstream analyses, unveiling functional cellular heterogeneity at an unprecedented scale.

About the speaker: Dr Shang is a tenure-track Assistant Professor in the Department of Biostatistics at MD Anderson Cancer Center. She obtained her PhD degree from the Department of Biostatistics at the University of Michigan. She is primarily interested in developing effective and efficient statistical and computational methods for analyzing large-scale genetic and genomic datasets. Specifically, her focus is on integrating multi-omics data in single-cell and spatial transcriptomics, ultimately connecting molecular insights with clinical applications.

August 26 - A maximum penalized likelihood approach for semiparametric accelerated failure time models with time-varying covariates and partly interval censoring

Speaker: Dr Aishwarya Bhaskaran (Macquarie University)

Abstract: Accelerated failure time (AFT) models are frequently used for modelling survival data. This approach is attractive as it quantifies the direct relationship between the time until an event occurs and various covariates. It asserts that the failure times experience either acceleration or deceleration through a multiplicative factor when these covariates are present. While existing literature provides numerous methods for fitting AFT models with time-fixed covariates, adapting these approaches to scenarios involving both time-varying covariates and partly interval-censored data remains challenging. In this paper, we introduce a maximum penalized likelihood approach to fit a semiparametric AFT model. This method, designed for survival data with partly interval-censored failure times, accommodates both time-fixed and time-varying covariates. We utilize Gaussian basis functions to construct a smooth approximation of the nonparametric baseline hazard and fit the model via a constrained optimization approach. To illustrate the effectiveness of our proposed method, we conducted a comprehensive simulation study. We also present an implementation of our approach on a randomized clinical trial dataset on advanced melanoma patients.

About the speaker: Dr Aishwarya Bhaskaran earned her PhD from the University of Technology Sydney, where she was supervised by Professor Matt Wand. Her PhD research focused on likelihood theory and methods for generalized linear mixed models. Currently, she is a postdoctoral research fellow at Macquarie University, where she specializes in analyzing high-performance predictive models using semi-parametric survival regression.

August 19 - 3D Cells Motion Analysis for Investigating Apoptotic Cell Clearance in Germinal Centres

Speaker: Dr Xufeng Lin (Garvan Institute)

Abstract: This presentation explores the application of advanced 3D cell motion analysis techniques to investigate apoptotic cell clearance within germinal centers (GCs) of lymph nodes, a process potentially linked to systemic lupus erythematosus (SLE) pathogenesis. Employing cutting-edge 3D cell segmentation, tracking, and classification methods, we aim to gain unprecedented insights into the behaviors and interactions of cells within GCs, focusing on tingible body macrophages (TBMs) and apoptotic GC B cells. By visualizing and quantifying the dynamics of cell motion, we seek to elucidate the relationship between defective clearance and loss of B cell tolerance. This research aims to advance our understanding of SLE pathogenesis and showcase the effectiveness of 3D cell motion analysis in unraveling complex cellular processes in immune responses.

About the speaker: Dr Xufeng Lin is a Senior Computational Scientist in the Computational Biology Group, Data Science Platform at Garvan Institute of Medical Research. He received his PhD in Computer Science in 2017 and previously held positions as a Research Fellow in Computer Vision at Deakin University (2019-2022) and an Associate Lecturer in Computing at Charles Sturt University (2017-2019). His research interests encompass Computer Vision, Machine Learning, and Multimedia Forensics. Dr. Lin has authored one book chapter and over twenty papers in top-tier journals and international conferences. He holds two patents and received the Best Paper Award at the 15th IEEE International Conference on Advanced Video and Signal-based Surveillance in 2018. Actively engaged in international collaborations, he was a Marie Curie Research Fellow at South China University of Technology in 2016, supported by the EU Horizon 2020 IDENTITY project, and a Visiting Researcher at the University of Salzburg in 2015, funded by the European Cooperation in Science and Technology (COST) Action IC1106.

August 12 - Scalable analysis methods for single cell and spatial omics data and lineage tracing

Speaker: Dr Joshua Ho (University of Hong Kong)

Abstract: Single cell (scRNA-seq) and spatial transcriptomic (ST) analyses are gaining widespread adoption in many areas of biomedical research. A large amount of scRNA-seq and ST data are being generated at a rapid rate. Nonetheless, it is challenging to quickly and efficiently process a large collection of these data. Extending our team’s pioneering experience in developing scalable bioinformatics tools, we have developed a suite of cloud-accelerated scRNA-seq analysis tools that efficiently process read-level information at a highly scalable manner. In this talk, we will showcase a number of tools that we have developed to enable large-scale single cell RNA-seq analysis. Furthermore, we will discuss new methods that enable somatic genetic variants in mitochondria to be used as endogenous lineage tracing markers in scRNA-seq and ST data. We will showcase the use of this type of novel lineage tracing technology to study cancer using human clinical tumour samples.

About the speaker: Joshua Ho is an Associate Professor in the School of Biomedical Sciences at the University of Hong Kong (HKU). He completed his BSc (Hon 1, Medal) and PhD in Bioinformatics from the University of Sydney, and undertook postdoctoral training in at Harvard Medical School. He is an expert in developing scalable computational methods to analyse single cell omic data, metagenomic data, and digital health data. Joshua has over 140 publications, including first or senior papers in leading journals such as Nature, Nature Communications, Genome Biology, Nucleic Acids Research and Cell Systems. Joshua is also currently the Programme Co-Director of BSc (Bioinformatics), Deputy Director of EdTech (AI), Academic Lead of the Bioinformatics Core of the Centre for PanorOmic Sciences (CPOS) at HKU, and Lead Scientist of Laboratory Data Discovery for Health. He co-founded a digital health company Vitome Limited.

August 5 - Using statistical genomics to identify sex-specific risk factors of cardiovascular diseases

Speaker: Dr Jiayue-Clara Jiang (University of Queensland)

Abstract: Current management and care policies for cardiovascular diseases (CVD) are mostly based on male-oriented evidence, and female-specific risk factors of CVDs remain poorly understood. To improve the understanding of female-specific CVD risk factors, we applied statistical genomic methods to investigate the sex-specific association between psychiatric disorders and the incident risks of CVDs in a cohort of ~345,000 European-ancestry individuals (UK Biobank). The genetic risks of three psychiatric disorders (major depression [MD], schizophrenia and bipolar disorder) were estimated using polygenic risk scores.

Cox proportional hazards regression analysis showed that having a higher genetic risk of MD was associated with increased incident risks of atrial fibrillation (AF), coronary artery disease (CAD) and heart failure (HF) in females, but not in males. These associations remained in the absence of any psychiatric disorder diagnosis or medication use, and were thus not simply a consequence of diagnosis or psychiatric medications. Mediation analysis showed that traditional risk factors did not explain the sex differences. We did not observe any association with CVD risks in either sex for the genetic risk of schizophrenia or bipolar disorder.

In conclusion, our findings highlight the sex differences in the risk profiles of CVDs, and show that a higher genetic risk of MD is a greater risk factor for CVDs in females versus males. Our findings warrant further investigation to understand the biological mechanisms mediating this sex difference. Further studies are essential to determine whether the genetic risk of MD is useful for improving the prediction of CVD risks, especially for females.

About the speaker: Dr Jiayue-Clara Jiang is a postdoctoral research fellow at the Institute for Molecular Bioscience, The University of Queensland (UQ). Clara's research focuses on using genomic and transcriptomic analysis to investigate the genetic basis of cardiovascular and psychiatric disorders, with a particular focus on female health, as well as using statistical genomic approaches to explore possible opportunities for drug repurposing. Clara graduated from the University of Queensland with Bachelor of Advanced Science (First Class Honours) in 2017, and was awarded the University Medal. Clara was awarded her PhD at the University of Queensland in 2021, where she utilised bioinformatic approaches and molecular experiments to decipher the genetic aetiology of breast cancer, specifically the regulatory role of transposons or 'jumping genes' in modulating the transcriptional landscape in the cancer state. Clara is also a UQ Wellness ambassador and an advocate for promoting equity, diversity and inclusion in academia.

June 3 - What was old is new again: anecdotes from the analysis of new single cell technologies

Judith and David Coffey Seminar

Speaker: Prof Alicia Oshlack (Peter MacCallum Cancer Centre)

Abstract: Single cell RNA sequencing technologies have been available for a decade. However, new technologies are still being developed and refined to provide deeper and more accurate insights into single cells. New technologies such as the chromium compatible 10X FLEX are designed to help with clinical sample collection and use a probe based capture design for RNA. We are using this technology to develop a cell atlas of the pediatric airways. Along the way we have been exploring aspects of design and analysis for this technology. This includes low level analysis and QC of cells as well as probe design and gene summarisation. Other new technologies include using long-read sequencing to explore full-length isoform in single cells. We are working on approaches to simulate and analyse differential transcript usage across single cells.

About the speaker: Professor Alicia Oshlack has been at the forefront of bioinformatics research for nearly 20 years. She is the Head of the Computational Biology Program and group leader at the Peter MacCallum Cancer Centre. She is best known for her large body of work on transcriptional analysis and medical genomics. In addition, Oshlack is internationally recognised for her development of bioinformatics methods for a range of applications including single cell RNA-seq, methylation and genomic analysis. Oshlack is involved in many cutting edge collaborative projects utilising high throughput sequencing to investigate disease and development. She has published more than 120 papers and has developed more than a twenty software packages. Oshlack has been recognised by several awards including the Australian Academy of Science, Gani Medal for Human Genetics (2011), the Georgina Sweet Award for Women in Quantitative Biomedical research (2016), Senior Fellow award from the Australian Bioinformatics and Computational Biology Society (ABACBS, 2023) and was recently elected as a member of the Australian Academy of Health and Medical Science.

May 27 - Connecting the brain with body in neuroimaging research and its implication in neuropsychiatry

Speaker: Dr Ye Ella Tian (University of Melbourne)

Abstract: Integrated research into brain and body systems holds substantial clinical potential in addressing multimorbidity and physical illness burden in people with neuropsychiatric disorders. In the first part of my talk, I will introduce a multiorgan aging network to demonstrate how the aging of one organ system selectively and characteristically influences the aging of other brain and body systems. I will show multiorgan aging profiles for 16 chronic brain and body disorders and its relationship to mortality risk prediction. Paper link: https://www.nature.com/articles/s41591-023-02296-6

In the second part of my talk, I will present a multiorgan, system-wide characterization of brain and body health for common neuropsychiatric disorders. I will show that individuals diagnosed with these neuropsychiatric disorders are not only characterized by deviations from normative reference ranges for brain phenotypes but also present considerably poorer physical health across multiple body systems compared to their healthy peers. I will show that poor physical health is a more pronounced manifestation of neuropsychiatric illness compared to brain changes. To close the talk, I will call for integrated and holistic mental and physical health care in psychiatry. Paper link: https://jamanetwork.com/journals/jamapsychiatry/fullarticle/2804355

About the speaker: Dr Ye Ella Tian is an NHMRC Emerging Leadership Fellow at the Department of Psychiatry, The University of Melbourne. She is a psychiatrist and neuroscientist by training and holds a PhD in systems neuroscience. She works at the interface between neuroscience, computation and translational research of applying brain imaging techniques to clinical research. She leads the development of the Melbourne Subcortex Atlas. Her current research focuses on brain-body relationships in mental illness across the lifespan.

May 20 - A statistical approach for removing joint and individual unwanted variation from single-cell multi-omics data

Speaker: Hsiao-Chi Liao (University of Melbourne)

Abstract: Single-cell multimodal technologies provide an opportunity to study biological mechanisms in a more comprehensive manner. CITE-seq (cellular indexing of transcriptomes and epitopes) assay simultaneously measures mRNA and surface proteins at the single-cell level, and is one of the most popular single-cell multi-omics platforms. The integrated analysis of mRNA expression and protein abundance can help reveal biological insight that would not have been possible from separate analyses of each modality. Unwanted variation from sources such as shared batches and domain-specific library size effects inevitably exists in data from both domains. If not properly corrected, the unwanted variation can potentially lead to misleading conclusions being made from the downstream analyses. We propose a method for removing unwanted variation from matched single-cell multi-omics data that allows us to estimate joint and modality-specific unwanted effects. In our preliminary study with four matched single-cell multi-omics datasets, we have shown that our approach is generally competitive in terms of preserving biological signals and mitigating the undesired technical effects compared to current methods such as Seurat, and can do better when the biological and unwanted variation are associated, as it can avoid removing too much biological signal from the data.

About the speaker: Hsiao-Chi is a doctoral candidate at the School of Mathematics and Statistics, The University of Melbourne, supervised by Dr. Agus Salim, Dr. Terry Speed, and Dr. Davis McCarthy. Her research interests are integrative analysis of single-cell multi-omics datasets and methods development for analysing omics data. She is currently working on her PhD projects that aim to develop statistical methods for removing unwanted variation from proteomics and transcriptomics data.

May 13 - Genetics of sensory nutrition

Speaker: Dr Daniel Hwang (University of Queensland)

Abstract: Perception of taste and smell shapes our food preferences and choices, playing a pivotal role in dietary habit. Given that dietary intake is a key risk factor of various chronic conditions, including obesity, cardioembolic disorders, and cancers, comprehending the impact of individual variations in sensory perception on eating behaviour and its subsequent effects on health is crucial. This presentation will delve into the genetics of taste perception and food intake, address current challenges and explore future directions for utilizing this knowledge in personalized nutrition and intervention strategies.

About the speaker: Daniel Hwang is an ARC DECRA Fellow at the Institute for Molecular Bioscience, The University of Queensland. He studied Biochemistry as an undergraduate at the National Taiwan University and a Master in Biotechnology at the University of Pennsylvania. Following graduation, he conducted research at the Monell Chemical Senses Center where he first developed a keen interest in genetics and the perception of smell and taste (See his taste work in The Conversation). He was later awarded scholarships to complete a master’s degree in Nutrition at the University of Washington and a PhD degree in Genetic Epidemiology at the QIMR Berghofer Medical Research Institute. Daniel develops and applies statistical methodologies to large-scale high-dimensional data to understand how genes influence human sensory perception, dietary behaviour and related health conditions. He is an EMCR member of the National Committee for Nutrition of the Australian Academy of Science and a Leadership Team member of the Global Consortium for Chemosensory Research. He is on the editorial boards of BMC Medicine and Twin Research and Human Genetics. A complete list of work can be found at: https://researchers.uq.edu.au/researcher/20291.

May 6 - Bacterial-host interactions in the context of the tumor microenvironment

Speaker: Dr Jorge Galeano (Fred Hutchinson Cancer Center)

Abstract: The tumor-associated microbiota has been gaining significant attention due to its ability to promote cancer progression in tumors from the gastrointestinal tract. Microbiome analysis has identified the microorganisms that could be associated with cancer cell malignancy during metastasis and chemoresistance. For instance, the enrichment of Fusobacterium nucleatum in the tumor tissue has been associated with poor clinical outcomes in patients with colorectal cancer. However, it is still unknown how these microorganisms are spatially distributed across the tumor tissue and with what elements of the tumor tissue they interact with to promote cancer progression. In this work, by modifying existing technologies that can map the spatial distribution of RNA transcripts and protein molecules along the tumor tissue, we found that intratumoral bacteria reside in distinct microniches that are functional different from other microcompartments from the same tumor sample. The bacteria-infected microniches were characterized to be hypoxic and largely immunosuppressive with an increased infiltration of pro-inflammatory myeloid cells such as neutrophils and macrophages and excluded from T cell. Cancer epithelial cells that resided in microniches containing bacteria exhibited limited capacity to proliferate with sever chromosome instability. We conclude that intratumoral bacteria is not randomly distributed across the tumor tissue, instead they are highly organized in distinct microniches that can modulate the biological funtion of other elements of the tumor microenvironment including the anti-tumor immune response and cancer epithelial cell compartments.

About the speaker: Jorge completed his medical degree at the National University of Colombia. Then he migrated to Sydney, Australia where he studied actin dynamics in cytotoxic T cells during antigen engagement with cancer cells at the University of Sydney. During his doctoral degree he investigated the molecular mechanisms that drive T cell recruitment in the tumor tissue at the EMBL Australia UNSW node. He is currently doing his postdoctoral training at the Fred Hutch, Seattle USA, where he is studying the influence of bacteria in promoting cancer development.

April 29 - Micro-macro adventures in cellular diversity

Speaker: Dr Fabio Zanini (UNSW)

Abstract: Single cell sequencing has recently opened a new frontier in biomedicine, enabling characterisation of cell identities and behaviours with granularity and scale. Most analyses take place at a mesoscopic scale of sorts: thousands of cells from a specific organ and organism (e.g. adult human blood) are individually sequenced and later computationally grouped (clustering, pseudotime) to gain biological insights. In this talk, I will tell two stories that challenge this paradigm in quite opposite ways. First, I will describe HyperSeq, a new experimental method combining microscopy and transcriptomics to find cellular diversity where there appears to be none: a healthy cell line. Second, I will outline atlas approximations, a desperate quest to fit all cellular diversity on Earth into a single online place.

About the speaker: Fabio Zanini is a group leader at UNSW focussing on single cell biomedicine and open source software development. After a PhD at the Max Planck Institute (Germany) and a postdoc at Stanford University (USA), he moved to Australia in 2019 to create the Data Driven Biomedicine lab (https://fabilab.org). His research covers three areas: (i) Data science on specific biomedical systems including development of the lung, viral infections, immunology, and cancer, (ii) Bioinformatic software including HTSeq 2.0 and igraph, and (iii) Innovation in single cell methods including computational algorithms and experimental protocols. Fabio's overall research passion is cellular diversity.

April 22 - Systematic comparison of sequencing-based spatial transcriptomic methods with cadasSTre and SpatialBench

Judith and David Coffey Seminar

Speaker: Prof Matthew Ritchie (WEHI)

Abstract: Sequencing-based Spatial Transcriptomics (sST) allows gene expression to be measured within complex tissue contexts. Although a wide array of sST technologies are currently available to researchers, efforts to comprehensively benchmark different platforms are currently lacking. The inherent variability across technologies and datasets poses challenges in formulating standardized evaluation metrics. To address this, we established a collection of reference tissues and regions characterized by well-defined histological architecture and other biological ground truth and used them to generate the cadasSTre and SpatialBench datasets that compare 11 sST methods. We highlight molecular diffusion as a variable parameter across different methods and tissues, significantly impacting the effective resolution. Furthermore, we observed that spatial transcriptomic data demonstrate unique attributes beyond merely adding a spatial axis to single-cell data, including an enhanced ability to capture patterned rare cell states along with specific markers, albeit being influenced by multiple factors including sequencing depth and resolution. For the 10X Visium platform, we benchmarked the performance of different sample handling approaches after preprocessing, explored spatially variable gene detection and the ability of clustering and cell deconvolution to identify expected cell types and tissue regions. Multi-sample differential expression analysis was able to recover known gene signatures related to biological sex or gene knockout. Our datasets and analyses serve as a practical guide for sST users and will be useful in future benchmarking studies.

About the speaker: Professor Matt Ritchie has been at lab head at the WEHI for the past 11 years. His team develops analysis methods and open-source software tailored to new applications of genomic technology in biomedical research. In the single-cell and spatial biology field, this work includes tools for data preprocessing (scPipe), benchmarking at scale (CellBench) and new protocols and analysis methods (FLAMES) for applying long-read sequencing to single-cell research. His most recent research is on developing benchmarking resources for sequencing-based spatial transcriptomics technologies (cadasSTre and SpatialBench). Matt completed his PhD on microarray data analysis at WEHI in 2005 under the supervision of Professor Gordon Smyth, which was followed by a period of post-doctoral research at the EBI (Hinxton, UK) and University of Cambridge before returning to WEHI as a Senior Research Officer in 2008. He is a keen advocate of open-source software, having served on both the Technical Advisory Board and Community Advisory Board of the Bioconductor project.

April 15 - High dimensional tensor methods for multi-modal single cell genomics data

Speaker: Kwangmoon Park (University of Wisconsin-Madison)

Abstract: Emerging single cell technologies that simultaneously capture long-range interactions of genomic loci together with their DNA methylation levels are advancing our understanding of 3D genome structure and its interplay with the epigenome at the single cell level. While methods to analyze data from single cell high throughput chromatin conformation capture (scHi-C) experiments are maturing, methods that can jointly analyze multiple modalities with scHi-C data are lacking. In this talk, I present two tensor modeling frameworks: Muscle and SHOPS, to jointly analyze 3D conformation and DNA methylation data measured at the single cell level. First, I present Muscle, a joint decomposition of Multiple single cell tensors. Muscle is a novel tensor decomposition method that can integrate the scHi-C and DNA methylation modalities with a direct interpretability. Next, I introduce SHOPS, Sparse Higher Order Partial Least Squares, which provides an inference on the direct association between Hi-C and DNA methylation. SHOPS is a new tensor response regression method to simultaneously achieve denoising of the scHi-C tensor and selecting the most relevant methylation sites with dimension reduction.

About the speaker: Kwangmoon Park is a Statistics Ph.D. Candidate at the University of Wisconsin-Madison. He is currently working on statistical genomics and high dimensional statistics with Professor Sündüz Keleş. Before joining UW-Madison, he earned a master’s degree in Statistics at the Yonsei University in 2020. He earned a B.A. in Economics and Statistics at Yonsei University in Korea and studied Economics as an exchange student at Erasmus Universiteit Rotterdam in the Netherlands. Kwangmoon Park is mainly interested in questions related to understanding how genes are regulated by distal regions in the genome, particularly by functional non-coding regions. For that purpose, he develops statistical tools for analyzing High-dimensional genomic data, including Hi-C and HiChIP, and for linking diverse types of genomic or epigenomic data with better statistical interpretation. The statistical methodologies he works on are related to tensor factorization/regression and dimension reduction techniques, including Partial Least Squares.

April 8 - scNovel: a scalable deep learning-based network for novel rare cell discovery in single-cell transcriptomics

Speaker: Yixuan Wang (CUHK)

Abstract: Single-cell RNA sequencing has achieved massive success in biological research fields. Discovering novel cell types from single-cell transcriptomics has been demonstrated to be essential in the field of biomedicine, yet is time-consuming and needs prior knowledge. With the unprecedented boom in cell atlases, auto-annotation tools have become more prevalent due to their speed, accuracy, and user-friendly features. However, existing tools have mostly focused on general cell type annotation and have not adequately addressed the challenge of discovering novel rare cell types. In this work, we introduce scNovel, a powerful deep learning-based neural network that specifically focuses on novel rare cell discovery. By testing our model on diverse datasets with different scales, protocols, and degrees of imbalance, we demonstrate that scNovel significantly outperforms previous state-of-the-art novel cell detection models, reaching the most AUROC performance(the only one method whose averaged AUROC results are above 94%, up to 16.26% more comparing to the second-best method). We validate scNovel's performance on a million-scale dataset to illustrate the scalability of scNovel further. Applying scNovel on a clinical COVID-19 dataset, three potential novel subtypes of Macrophages are identified, where the COVID-related differential genes are also detected to have consistent expression patterns through deeper analysis. We believe that our proposed pipeline will be an important tool for high-throughput clinical data in a wide range of applications.

About the speaker: Yixuan Wang is a second-year Ph.D. student in the Department of Computer Science and Engineering at The Chinese University of Hong Kong, co-advised by Prof. Yu Li and Prof. Irwin King. She received her B.S. degree at the Harbin Institute of Technology in 2022. She focuses on developing innovative deep learning approaches to address computational issues in the realms of biology and healthcare, with a specific emphasis on tackling challenges related to single-cell data. She has published six papers in Nature Communications, Bioinformatics, Briefings in Bioinformatics, and RECOMB.

March 25 - Benchmarking long-read RNA-sequencing analysis tools using in silico mixtures

Speaker: Dr Xueyi Dong (WEHI)

Abstract: The lack of benchmark datasets with inbuilt ground-truth makes it challenging to compare the performance of existing long-read isoform detection and differential expression analysis workflows. Here, we present a benchmark experiment using two human lung adenocarcinoma cell lines that were each profiled in triplicate together with synthetic, spliced, spike-in RNAs (“sequins”). Samples were deeply sequenced on both Illumina short-read and Oxford Nanopore Technologies long-read platforms. Alongside the ground-truth available via the sequins, we created in silico mixture samples to allow performance assessment in the absence of true positives or true negatives. Our results show that StringTie2 and bambu outperformed other tools from the 6 isoform detection tools tested, DESeq2, edgeR and limma-voom were best amongst the 5 differential transcript expression tools tested and there was no clear front-runner for performing differential transcript usage analysis between the 5 tools compared, which suggests further methods development is needed for this application.

About the speaker: Dr Xueyi Dong is a postdoctoral research officer in Chen lab in ACRF Cancer Biology and Stem Cells division, the Walter and Eliza Hall Institute of Medical Research (WEHI). She did her undergraduate in Zhejiang University in China, majored in biology science (2014-2018). She completed her PhD at WEHI in 2023 under the supervision of Prof. Matthew Ritchie, Dr. Charity Law and Prof. Gordon Smyth. Her current research primarily involves the analysis of spatial transcriptomics data and the investigation of RNA splicing.

March 18 -  Leveraging genetic diversity to fine-map causal variants of complex traits

Speaker: Dr Mingxuan Cai (CUHK)

Abstract: Fine-mapping prioritizes risk variants identified by genome-wide association studies (GWASs), serving as a critical step to uncover biological mechanisms underlying complex traits. The major challenges of fine-mapping arise from the homogeneous LD patterns and unadjusted confounding bias in GWAS samples, leading to sub-optimal power and false positives. Here, we develop a statistical method for cross-population fine-mapping (XMAP) by leveraging genetic diversity and accounting for confounding bias. By using cross-population GWAS summary statistics from global biobanks and genomic consortia, we show that XMAP can achieve greater statistical power, better control of false positive rate, and substantially higher computational efficiency for identifying multiple causal signals, compared to existing methods. Importantly, we show that the output of XMAP can be integrated with single-cell datasets, which greatly improves the interpretation of putative causal variants in their cellular context at single-cell resolution.

About the speaker: Dr. Cai is an Assistant Professor at Department of Biostatistics, City University of Hong Kong. He obtained his PhD degree from The Hong Kong University of Science and Technology in 2022. His broad area of interest lies in statistical machine learning and data science with applications in genetics and genomics data analysis.

March 11 - From genetic variants to cellular function: morphological profiling genes associated with atrial fibrillation

Speaker: Dr Ling Xiao (Harvard Medical School)

Abstract: Atrial fibrillation (AF) is a complex disease and the molecular mechanisms leading to AF in the general population remain unknown. Genome-wide association studies (GWAS) identified over 100 genetic loci associated with AF. However, it is still a big challenge to identify cellular programs through which genes from AF-associated variants and genetic loci modulate the risk for AF. To address this challenge, we systematically applied high-content imaging assay to analyze AF genes in a human pluripotent stem cell (hPSC) derived atrial cardiomyocyte (aCMs) cell model. We performed CRISPR based perturbations to delete AF candidate genes in Cas9 expressing hPSC derived aCMs and used Cell Painting pipeline to analyze atrial cell structures obtained with high-content imaging. Our results indicate cardiomyocyte structural abnormalities contribute to AF pathogenesis. Understanding the functions of candidate AF genes at the GWAS loci have the potential to uncover novel biological mechanisms and potential drug targets for novel therapeutics.

About the speaker: Dr. Ling Xiao is an Instructor in investigation at Cardiovascular Research Center of Massachusetts General Hospital and an Instructor in Medicine at Harvard Medical School. Her research focuses on leveraging large human genetic dataset and state-of-the-art functional genomic approaches to study cardiovascular diseases.

March 4 - Leveraging natural genetic variation to understand human biology  

Judith and David Coffey Seminar

Speaker: Prof Daniel MacArthur (Garvan Institute)

Abstract: The human population, through explosive growth, has performed a comprehensive saturation mutagenesis experiment on itself: any single base substitution that is compatible with life is expected to be present somewhere in the genome of at least one of the nearly 8 billion living humans. Our species has thus, in effect, done many of the natural experiments required to understand our own genotype-phenotype map; the goal of geneticists is to generate the right data from the right people to understand this map, and to convert it into actionable information that can be used in the prediction, diagnosis, and treatment of disease. In this presentation I will discuss the impact of large-scale genomics in international biobanks and healthcare settings on our understanding of human biology; review some of the most important goals for the field of human genetics over the next 5 years; and emphasise the urgent need for more representative resources of human genetic and genomic data to ensure the equitable benefit of future advances in genomic medicine.

About the speaker: Daniel is the Director of the Centre for Population Genomics, based jointly at the Garvan Institute of Medical Research, Sydney and the Murdoch Children’s Research Institute in Melbourne.

He completed his PhD at the University of Sydney before moving to postdoctoral studies at the Wellcome Trust Sanger Institute in Cambridge, UK, and then a faculty position at Harvard Medical School, Massachusetts General Hospital, and the Broad Institute of MIT and Harvard in Boston. In this position he co-directed the Broad Institute’s Program in Medical and Population Genetics, as well as the NIH-funded Center for Mendelian Genomics, which sequenced the exomes, genomes, and/or transcriptomes of over 10,000 individuals from families affected by severe Mendelian disease. He also led the Genome Aggregation Database (gnomAD) consortium, which produces the world’s largest publicly accessible catalogue of human genetic variation, now spanning data from more than 800,000 individual exomes and genomes.

Daniel returned to Australia in 2020 to lead the new Centre for Population Genomics (CPG), now a team of 40 researchers, software engineers, community engagement experts, and other professional staff. The Centre's mission is to establish respectful partnerships with diverse Australian communities, to work with those communities to collect and analyse genomic data at transformative scale, and to use these data to drive both novel genomic discovery and the development of equitable genomic medicine. The CPG currently leads national projects in the development of more representative resources of genetic variation in Australian communities; improving the genomic diagnosis of rare disease; and combining large-scale genetic and cellular genomic data to understand gene function.

February 26 - Single-cell and Spatial Transcriptomics Clustering with an Optimized Adaptive K-Nearest Neighbor Graph

Speaker: Dr Jia Li (Vanderbilt University Medical Center) 

Abstract: Single-cell and spatial transcriptomics have been widely used to characterize cellular landscape in complex tissues. To understand cellular heterogeneity, one essential step is to define cell types through unsupervised clustering. While typical clustering methods have difficulty in identifying rare cell types, approaches specifically tailored to detect rare cell types gain their ability at the cost of poorer performance for grouping abundant ones. Here, we developed aKNNO, a method to identify abundant and rare cell types simultaneously based on an adaptive k-nearest neighbor graph with optimization. Benchmarked on 38 simulated and 20 single-cell and spatial transcriptomics datasets, aKNNO identified both abundant and rare cell types accurately. Without sacrificing performance for clustering abundant cell types, aKNNO discovered known and novel rare cell types that those typical and even specifically tailored methods failed to detect. aKNNO, using transcriptome alone, stereotyped fine-grained anatomical structures more precisely than those integrative approaches combining expression with spatial locations and histology image. 

About the speaker: Dr Jia Li is currently a Postdoctoral Fellow in the Department of Biostatistics in Vanderbilt University Medical Center. Her research is focused on the analysis and method development for single cell RNA sequencing and spatial transcriptomics data. 

October 30 - Integrative annotation scores of variants for impact on RNA binding protein activities

Speaker: Jingqi Duan (University of Wisconsin-Madison)

Abstract: The ENCODE project generated a large collection of eCLIP-seq RNA binding protein (RBP) profiling data with accompanying transcriptome data from RNA-seq experiments of RBP knockdowns by shRNA. However, these datasets are not fully exploited to elucidate the impact of genetic variants on RBP activities. We implement INCA (Integrative annotation scores of variants for impact on RBP activities) as a multi-step genetic variant scoring approach that leverages the ENCODE RBP data together with ClinVar and integrates multiple computational approaches to aggregate evidence. INCA hinges upon evaluating the impact of the variants on the RBP activities by leveraging the genotyped cell lines that harbor these variants. We show that INCA provides critical specificity for the set of candidate variants and their linkage disequilibrium partners even after they are generically scored for impact on RBP binding. As a result, it can augment scoring of 46.2% of the candidate variants for follow-up on average.

About the speaker: Jingqi Duan is a fourth-year Ph.D. student in the Department of Statistics at the University of Wisconsin-Madison. She is currently a research assistant in the Keles Research Group. Her research focuses on advancing statistical and computation methods tailored for the analysis of high-throughput sequencing data, such as eCLIP-seq and Perturb-seq, with the aim of enhancing gene regulation analysis.

October 23 - Statistical and computational methods for spatial transcriptomics data analysis

Speaker: Dr Ying Ma (Brown University)

Abstract: Spatial transcriptomics technologies have enabled gene expression profiling on complex tissues with spatial localization information. The majority of these technologies, however, effectively measure the average gene expression from a mixture of cells of potentially heterogeneous cell types on each tissue location. Here, I develop a deconvolution method, CARD, that combines cell-type-specific expression information from single-cell RNA sequencing (scRNA-seq) with correlation in cell-type composition across tissue locations. Modeling spatial correlation allows us to borrow the cell-type composition information across locations, improving accuracy of deconvolution even with a mismatched scRNA-seq reference. CARD can also impute cell-type compositions and gene expression levels at unmeasured tissue locations to enable the construction of a refined spatial tissue map with a resolution arbitrarily higher than that measured in the original study and can perform deconvolution without a scRNA-seq reference. In a real data application on the human pancreatic ductal adenocarcinoma (PDAC) dataset, CARD identified multiple cell types and molecular markers with distinct spatial localization that define the progression, heterogeneity, and compartmentalization of pancreatic cancer. In addition, if time allows, I will also discuss my other methodological work on integrative differential expression and gene set enrichment analysis in scRNA-seq studies, integrative reference-informed tissue segmentation in SRT studies, and collaborative work on polygenic risk scores for common health-related exposure traits in the Michigan Genomics Initiative (MGI) cohort.

About the speaker: Dr. Ying Ma is an Assistant Professor at the Department of Biostatistics and the Center for Computational Molecular Biology at Brown University. Her research interests focus on developing efficient statistical learning methods to address a variety of biological problems and computational challenges in genomics and genetics. These challenges typically arise with the high-dimensional data generated by rapidly evolving sequencing technologies, e.g., single-cell RNA-seq (scRNA-seq), and spatially resolved transcriptomics (SRT). With the emergence of these large-scale data, she has been continually motivated to develop tailored statistical models to advance our understanding in cellular heterogeneity, tissue organization, and the underlying mechanisms of various types of cancers. Besides her genomics research, she also works on genetic risk prediction and polygenic risk score problems in large biobanks such as UKBiobank, and MGI.

October 9 - Modelling Host-Microbiome Interactions in Physiological and Pathological Processes

Judith and David Coffey Seminar

Speaker: Dr Elaine Holmes (Health Futures Institute, Murdoch University)

Abstract: The use of metabolic profiling to define metabolic phenotypes associated with a wide range of pathologies is expanding and demand for sensitive, high quality disease diagnostics has facilitated the development of new technological and statistical methods for extracting biomarkers from spectroscopic data obtained from biofluids such as urine, serum and stool extracts. These metabolite signatures can subsequently modelled with other ‘-omic’ data, including next generation sequencing data in order to establish connections between the gut bacteria and human (patho)physiology. Examples of urinary or faecal metabolites that are products of the microbiota, or microbiota-host interactions include phenols, indoles, bile acids, short chain fatty acids and choline derivatives, all of which can be quantitatively profiled using spectroscopic technology. Thus the metabolic phenotype can provide a window onto dynamic biochemical responses to physiological and pathological stimuli and also contains information relating to the metabolic activity and function of the gut microbiome.

In order to optimise information recovery from the spectra, analytical strategies for spectral alignment, scaling, curve resolution and quantification, statistical correlation and annotation are necessary. Some exemplar analytical pipelines are presented here with particular focus on a series of methods for enhancing biomarker detection via a family of statistical correlation algorithms. Cross-correlation of multiplatform data allows further characterisation and extraction of improved molecular descriptors of metabolites identified as candidate biomarkers, which in turn, can provide new insights into perturbed pathways and aetiopathogenetic mechanisms through correlation hierarchies of related metabolites. This systems analysis framework extends to encompass other datatypes such as metagenomic or metatranscriptomic data and can identify new correlates between datasets and establish biological coherence across metabolic pathways and networks.

About the speaker: Elaine Holmes is an ARC Laureate Fellow at Murdoch University, where she runs the Centre for Computational and Systems Medicine in the Health Futures Institute. She was elected as a Fellow of the Academy of Medical Sciences in 2018 and the Australian Academy of Science in 2022. Holmes is one of the pioneers in the development and implementation of metabolic phenotyping in translational clinical paradigms. The analytical framework conceptualised for metabolic phenotyping and biomarker discovery has been applied across several disease areas. She also co-developed the Metabolome-Wide Association Study concept and has shown that the microbial component of the metabolic profile is associated with a wide range of conditions including obesity, inflammatory bowel disease, allergies, and certain cancers. Her current focus is around computational modelling of metabolic and metagenomic data to understand the role of the gut microbiome in healthy aging with specific interest in the influence of nutrition on the microbiome.

September 25 - Statistical methods for complex kidney disease data 

Speaker: Dr Yunwei Zhang (Macquarie University)

Abstract: This talk gives an accessible and intuition-based overview of my PhD project, "Statistical methods for complex kidney disease data". I will start with highlighting some of the useful statistical modelling and data science techniques within the kidney disease domain. Motivated by improving the personalised healthcare for end-stage kidney disease patients, I will walk you through my journey of tackling the statistical challenges that arise when simulating deceased kidney donor allocation approaches, predicting post-transplant all-type survival and devising visualisation tools for individualised decision-making. This talk highlights how applied statistical and data science methods make a real impact on addressing real-world health challenges.

About the speaker: Yunwei Zhang began her PhD in July 2019 under the supervision of Jean Yang and Samuel Muller. She submitted her thesis in December 2022 and has since moved into Postdoctorate roles at the University of Melbourne and now at Macquarie University. She is interested in and works on methods that improve personalised healthcare in interdisciplinary fields by using statistics.

September 18 - Harnessing the power of artificial-intelligence for improved MRI-guided cancer radiotherapy 

Speaker: Dr David Waddington (The University of Sydney)

Abstract: Real-time tumour targeting with MRI-guidance has recently become possible with the advent of the MRI-Linac, which combines the unrivalled image quality of MRI with a linear accelerator (Linac) for x-ray radiation therapy. However, the relatively low spatio-temporal resolution of real-time MRI reduces the accuracy with which radiation beams can be adapted to tumour motion (e.g. respiration). New low-latency imaging techniques are thus essential to improving the quality of MRI-Linacs treatments. In this seminar, I will describe artificial-intelligence-based tools we have developed that can improve target tracking during radiation treatments. I will discuss the successes we have had ( and the challenges  faced) in deploying these AI tools to clinical systems.

About the speaker: Dr David Waddington is an NHMRC Emerging Leadership Fellow at the Image X Institute in the Faculty of Medicine and Health at the University of Sydney. As an early career researcher (PhD(Science), The University of Sydney, 2018), he specializes in developing new imaging technologies based on Magnetic Resonance Imaging (MRI) that can be used for the targeting of cancer therapeutics. David has published high impact, first-author research articles in multidisciplinary journals including Nature Communications and Science Advances that have led to two patent applications. He has been the recipient of prestigious academic awards including a 2013-14 Postgraduate Fulbright Scholarship (Harvard University) and a University Medal in Physics (UNSW - 2010). His work has won international conference prizes including two Best in Physic sat the American Association of Physicists in Medicine (AAPM - 2020, 2022) and two Summa Cum Laude awards at the International Society of Magnetic Resonance in Medicine (ISMRM - 2016, 2020). David has given several invited talks at international imaging conferences and presented at public events such as TEDx.

September 4 - Understanding cancer in adolescents and young adults through clinical panel sequencing 

Speaker: Dr Siyuan Zheng (UT Health, San Antonio)

Abstract: Cancer poses unique challenges in adolescents and young adults (AYAs) for several reasons. First, because of their relatively young age, they are often diagnosed late. Second, AYAs often face reproductive and developmental related issues that are usually of less concerns for older cancer patients. Third, long-term health problems associated with cytotoxic treatments can cause morbidity affecting decades of patients’ life. Forth, for unknown reasons, AYAs are less enrolled in clinical trials in the US. Finally, research attentions are more directed to pediatric and older adult patients. Many cancer hospitals have not yet established AYA specific clinical care programs. To improve our understanding of AYA cancers, we analyzed clinical panel sequencing data from AACR GENIE. GENIE is a consortium effort that comprise sequencing data for more than 100,000 patients. The dataset is highly heterogeneous, but meanwhile represents a rich resource for genomic studies due to its sample size and diversity. In this presentation, I will discuss our findings on the genomic and clinical disparities of AYA cancers.

About the speaker: Dr. Zheng is an Assistant Professor and PI at Greehey Children’s Cancer Research Institute, the University of Texas Health Science Center at San Antonio. He received his B.S. in Biochemistry from Nanjing University, China, and PhD in Bioinformatics from Shanghai Institutes for Biological Sciences, Chinese Academy of Sciences, China. After his PhD, he received postdoctoral training at Vanderbilt University and UT MD Anderson Cancer Center. During his time at MD Anderson, he participated in TCGA and led the TCGA adrenal cancer project. At UT Health, his group mines cancer genomic datasets to understand cancer aneuploidy, telomeres, and cancer disparity. He is a CPRIT Scholar in Cancer Research and a recipient of the UT Rising STARs Award in 2019. More information about him can be found at on his Github site.

August 28 - Polygenic risk score analysis with the addition of higher order interactions provides insight into protective and risk components of type 2 diabetes 

Speaker: Keri Multerer (Victoria University of Wellington)  
 
Abstract: Polygenic risk scores (PRS) based on genome wide association studies (GWAS) have been studied since about 2008. The PRS concept is based on obtaining individual DNA sequence information that is then compared for disease association using summary statics (derived from GWAS) in public databases. However, current PRS are not yet robust enough to be generally used in a clinical setting for flagging high risk individuals. Using data available in the UK Biobank with type 2 diabetes as the disease model, we have developed novel methods to incorporate higher order (epistatic) interaction weights to be included in PRS. This will help us to better understand if and by how much higher order interactions explain individual genetic risk. Currently we have improved feature selection for both main effect and epistatic interactions without sacrificing interpretability of the results leading to novel insights into the risk and protective variants driving type 2 diabetes.  
 
About the speaker: Keri Multerer is a 3rd year PhD student at Victoria University of Wellington in New Zealand under co-supervision of Andrew Munkacsi and Paul Atkinson. She earned a MSc in Genetics from George Washington University in Washington DC and worked in cancer research at Fred Hutchinson Cancer Research Center in Seattle before taking time off to raise three daughters. Teaching herself to code (in python) with specialised machine learning courses, Keri has integrated these two passions in her PhD thesis contributing to the fields of public health and personalised medicine. 

August 21 - Genome Canada Transplant Consortium (GCTC) computational approaches in kidney transplantation 

Speaker: Dr Oliver P. Günther (Günther Analytics; for the Genome Canada Transplant Consortium/Vancouver Immunology Lab)

Abstract: Chronic kidney disease is a major societal challenge, costing Canada over $30 billion/year. Transplantation can restore life-long health with tremendous cost savings, but antibody-mediated rejection (AMR) and complications of long-term immune suppression cause premature graft loss or early death in 50% of recipients, with increased costs of care. There are no proven treatments for AMR so prevention and early detection are crucial. Two Genome Canada Transplant Consortium (GCTC) projects will be presented: (1) Use of HLA genotyping and donor-recipient eplet-matching strategies in organ allocation simulation models, and (2) Longitudinal monitoring of gene expression in the peripheral blood of transplant patients to identify immune quiescence or activity. Results from simulations are used to inform organ allocation strategies involving eplet-based matching in Canada while results from the longitudinal genomics analysis will be used to inform AMR-monitoring strategies in combination with other data sets.

About the speaker: Dr Günther has worked with the Immunology Lab at Vancouver General Hospital/University of British Columbia, Canada, as a consultant for the past 5 years, providing customized data analysis, modeling and simulation for projects related to kidney transplantation. He received a Ph.D. in physics from the Goethe University Frankfurt, Germany in 1998. Since then he has focused on data analysis, visualization, modeling and simulation, as well as algorithm and software development, including positions at the University of British Columbia, the British Columbia Centre for Disease Control, and the Prevention of Organ Failure Centre of Excellence in Vancouver. Since 2014, Dr Günther provides consulting services through Gunther Analytics.August 14 - Learning consistent subcellular landmarks to quantify changes in multiplexed protein maps 

August 14 - Learning consistent subcellular landmarks to quantify changes in multiplexed protein maps

Speaker: Dr Scott Berry (UNSW)

Abstract: Highly multiplexed imaging holds enormous promise for understanding how spatial context shapes the activity of the genome and its products at multiple length scales. We have recently developed a deep-learning framework called CAMPA (Conditional Autoencoder for Multiplexed Pixel Analysis) to learn representations of molecular pixel-profiles that are consistent across heterogeneous cell populations and experimental perturbations. CAMPA identifies consistent subcellular landmarks, which can be quantitatively compared in terms of their sizes, shapes, molecular compositions, and relative spatial organisation. Using high-resolution multiplexed immunofluorescence, this reveals how subcellular organisation changes upon perturbation of RNA production, RNA processing, or cell size, and uncovers links between the molecular composition of membraneless organelles and cell-to-cell variability in bulk RNA synthesis rates. By capturing interpretable cellular phenotypes, we anticipate that CAMPA will accelerate the systematic mapping of multiscale atlases of biological organisation to identify the rules by which context shapes physiology and disease.

About the speaker: Scott has a background in Theoretical Physics and Molecular Biology. He studied a PhD at the John Innes Centre in Norwich, UK, on mechanisms of epigenetic memory in plants, before moving to the University of Zurich in Switzerland as an HFSP and EMBO postdoctoral fellow. In Zurich, Scott worked on mechanisms of mRNA concentration homeostasis in mammalian cells and experimental and computational approaches for acquiring and analysing highly multiplexed image data. In 2021, Scott started his group at Single Molecule Science, the EMBL Australia node at the University of New South Wales in Sydney. His group works on quantitative regulation of gene expression at the single-cell level, primarily employing microscopy and systems biology approaches – including mathematical modelling.