UHN Office of Research Trainees’ Post

In the realm of Bayesian Statistics, the utilization of both rbeta and rbinom is commonplace. The rbeta function finds its significance as a prior distribution, while rbinom assumes the role of the likelihood function. Through the application of Bayes' rule, the outcome emerges as a posterior distribution of the beta type. This formulation, recognized as the Beta-Binomial model, encapsulates the integration of rbeta and rbinom within Bayesian analysis. This model serves as a powerful tool for handling uncertainty and updating beliefs in various contexts, such as in clinical trials and quality control processes.

Learn from our panel of biostatisticians in next week's webinar as they discuss strategies for multivariable adjustment, the differences and appropriate applications of various regression analyses, and more! #EmergencyMedicine Register now: https://ow.ly/vPty50QoWhn

This ICES Research Forum 2024 Spotlight Session will provide an overview of the types and characteristics of Common Data Models, look at examples of Canadian studies undertaken with CDMs, and highlight new statistical methods for analysis of observational data developed using a CDM. Save your seat and register https://lnkd.in/gjECrCZY

Elephant in the room: prognostic covariates effect when leveraging external data. Gaining efficiency is important for trial design & analysis while maintaining validity! Our recent work on leveraging external evidence using Bayesian hierarchical model.

Today’s blog explores how single-cell self-contained hierarchical clustering (sc-SHC) incorporates model-based statistical testing during single-cell clustering. sc-SHC enhances single-cell analysis workflows by permitting statistical evaluation of clusters as distinct cell populations, providing confidence towards the discovery of novel cell types. Read about how the difficulty of reproducing clusters can be resolved with sc-SHC: https://lnkd.in/eCwJA246

The following decision tree shows how to handle missing values in observational studies, based on a paper by Kristin Sainani. This paper also provides guidance for dealing with missing values in randomized controlled trials and the special case of longitudinal studies. It is a highly recommended read for anyone interested in this topic. Sainani KL. Dealing With Missing Data. PM&R. 2015;7(9):990-994 [https://lnkd.in/eTEwsF59] #biostatistics

Today's blog focuses on a solution for one of the biggest problems in scRNA-seq data processing. What is the problem? Answer: Over-clustering. Yes, over-clustering and the subsequent spurious results can be horrible and you should be aware of it. However, a recent Nature paper by Grabski et al has a solution called sc-SHC, a novel algorithm that integrates statistical testing for accurate cell population identification and paves the way for reliable discovery of novel cell types - and sidesteps over-clustering. #bioinformatics #scRNA #overclustering

Following on from our recent ‘Prior Elicitation for Clinical Research’ webinar we’re pleased to share the recording and slide deck which can be downloaded here: https://lnkd.in/gh79Back. You will also be given access to additional resources that provide valuable insights from our expert Statisticians on the applications of Bayesian Statistics in clinical trials. #Bayesian #Phastar #Statistics

Alhamdulillah another Publication was done in HEC "Y" and PMDC recognized Journal with Medical College colleagues. #2023goals #2023healthtrends #statisticalanalysis #biostatistician #dataanalysis #researchmethodology #statistics #dataexperts

Always amazed at the amount of data you can generate by taking a blood sample and measurments. Very interested to see the data that can be generated from this research.