DCMB Tools & Technology Seminar - Oct 22, 2026 - Presented by Abhirath Anand @MichiganMedicine
DCMB Tools & Technology Seminar - Oct 22, 2026 - Presented by Abhirath Anand  @MichiganMedicine
Uploaded October 2026 | Updated November 2026, 3 days ago
"hospitraceR: An R Package for WGS-Based Genomic Epidemiological Analysis of Healthcare-Associated Infections”

Presented by Abhirath Anand
Department of Computational Medicine & Bioinformatics

Date: October 22, 2026

- Tool Link: github.com/theabhirath/hospitraceR

Abstract

Genomic epidemiology, which combines whole-genome sequencing (WGS) with patient data is becoming increasingly common to detect the transmission of hospital-acquired infections, especially multi-drug resistant organisms. As high-throughput WGS becomes more affordable, software that facilitates the analysis of this genomic data along with patient metadata becomes crucial to generate timely insights. Prior studies have shown that there can be many different approaches to detect transmission clusters with genomic epidemiology data, and oftentimes these approaches will have tradeoffs more or less suitable for a certain dataset. We present hospitraceR, an R package designed to facilitate such analyses. hospitraceR includes functions for detecting transmission clusters with two different methods, the standard SNV threshold approach and a threshold-free approach. It also includes functions that can use patient movement information through the facility to suggest potential transmission explanations for the clusters. hospitraceR seeks to enable users to use multiple methods of transmission cluster detection and compare them easily using cluster comparison metrics. Metrics like the adjusted Rand Index (ARI) and adjusted mutual information (AMI) have been included for this purpose, as has fraction of converts with shared source (FSS), a custom metric designed specifically for comparing transmission clusters. A companion package, hospitraceRvisualize, also includes opinionated plotting functions for certain commonplace visualizations that aid in epidemiological interpretations. We apply hospitraceR to a dataset consisting of 4 long-term acute care hospitals (LTACHs) from Chicago sampled for different periods of time as part of an intervention study and evaluate the two different approaches for cluster detection. We find distinct patterns exhibited by the transmission clusters in different locations and also compare the different clusters obtained with different SNV thresholds and linkage criterion for hierarchical clustering. The hospitraceR package proves to be helpful in facilitating such comparisons and makes it easy to use WGS data to aid epidemiological investigation.
DCMB Tools & Technology Seminar - Oct 22, 2026 - Presented by Abhirath AnandScintigraphy & SPECT-CT of the Parathyroid Glands: HyperparathyroidismComputational Medicine & Bioinformatics Seminar by Luis Zaman, PhD - Oct 14, 202620th Annual Prechter Lecture: Celebrating Legacy and the FutureHappy National Volunteer Week 2026!UM Health Peer Mentor Brenda BarkerDaTscan Imaging of Movement Disorders: A Nuclear Medicine CNS Imaging GuideCT of Diverticulitis: What to Know On CallNursing Professional Development Week 2025Tools for Youth Suicide Prevention: Understanding Michigans Firearm Safety LawsImaging of Cystic Lung Nodules in Lung Cancer Screening: Lung-RADSWhat is Lymphedema?
Michigan Medicine |

DCMB Tools & Technology Seminar - Oct 22, 2026 - Presented by Abhirath Anand

SHARE TO X SHARE TO REDDIT SHARE TO FACEBOOK WALLPAPER