BDSI Interactive Forum Series
Body
From the earliest days of statistics, data science and life science have been deeply intertwined. Biological data science carries this forward to a modern era in which the methods for collecting life science data are increasingly automated digital sensors. This deluge of data is transforming life science research across diverse scales and domains. Biological data science, broadly defined, is key to this transformation, from molecules to ecosystems, and across agriculture, medicine, biodiversity and bioengineering.
This interactive biological data science forum series aims to transect domains, connecting diverse researchers with a common interest in biological data science. The purpose is to invigorate interactions among us through an inclusive (fortnightly) forum that promotes the exchange of knowledge. We strive to define biological data science in the broadest sense, encompassing theory, methodology and applications as well as biological data acquisition, visualisation and analysis.
Forums held weekly from 2-3.30pm on Thursdays at Robertson #46 (DNA Room S104).
To register a short presentation, please contact BDSI
All welcome to attend.
Past events
This talk will present ongoing efforts to enhance sperm analysis using artificial intelligence, focusing on improving precision and consistency in motility and morphological assessments—key metrics in reproductive health studies.
Originally from the wild jungles of Borneo, David somehow found his way to becoming a research scientist at the Irradiation Immunity Interaction (i-Cube) Laboratory at the John Curtin School of Medical Research (JCSMR).
Elle will unveil the secrets behind her work on the gill microbiome of Atlantic Salmon, exploring its variation, heritability, and the implications of distance measures in her research.
The study seeks to analyse the fiber content of plant samples using two different filter bags, which yield different results due to particle size differences. F57 filter bags, intended for larger particles, result in significant particle loss and produce less reliable Acid Detergent Fiber (ADF) readings. F58 filter bags, designed for finer particles, provide more accurate and reliable ADF measures.
Masters Project. This research explores the potential of deep learning techniques to predict genetic mutations from Hematoxylin and Eosin (H&E) stained Whole Slide Images (WSIs) of tumor tissues.
Dr. Sehrish Kanwal will share her experience as a participant in the international rOpenSci Champions Program for 2024. Renowned for its cohort-based training, project development, and personalised mentorship, provides a platform for researchers to champion open science principles and practices.
Score matching, introduced by Hyvärinen (2005), is an estimation technique that circumvents the challenge of intractable normalizing constants in model densities.
Since the early twenty-first century, female life expectancy has consistently exceeded that of males worldwide.
Deep graph matching techniques have shown promising results in recent years. In this work, we cast deep graph matching as a contrastive learning task and introduce a new objective function for contrastive mapping to exploit the relationships between matches and non-matches.
Explore an overview on several teaching initiatives and development in Statistics at Macquarie University over the last couple of years, including ASA DataFest, the revamping of the Statistics Major in the Bachelor of Science (BSc) program and if time permits, the establishment of the Alumni Connection Lead role within the school.
DEPLOY, an integrated deep learning framework that predicts DNA methylation from histopathology images, and subsequently classifies brain tumours into 10 major subtypes.
Do you have some code or functions in R that you would like to make and distribute as an R package? Refactoring your R code as modular functions or organising your functions as an R package is a useful endeavour for scientific dissemination.