The project will involve the implementation and evaluation of algorithms for detecting transmission clusters in healthcare settings using genomic data.
[!WARNING] This project is still in the early stages of development and the API may change without notice.
Installation
transclust is not available on CRAN yet, but you can install it from GitHub using devtools:
devtools::install_github("theabhirath/transclust")Usage
Epidemiology data
We also provide epidemiological trace data for the isolates in the dataset. This is used in conjunction with the genomic data to identify clusters of transmission. The data is provided in an Rdata file, which can be loaded into R using the load() function. It has multiple data objects, including:
-
trace_mat: A matrix of trace data. Rows are days, columns are patients. -
floor_mat: A matrix tracking patient locations on the floor of the hospital. Rows are days, columns are patients. -
room_mat: A matrix tracking the room assignments of patients. Rows are days, columns are patients. -
dates: A vector of admission dates for each patient. -
dna_pt_labels: A vector mapping the sequence IDs for isolates to patient IDs. -
ip_seqs: A vector of sequence IDs which correspond to intake patient sequences presumed to be imported. -
ip_seqs_3days: A vector of sequence IDs which correspond to all intake patient sequences that tested positive within 3 days of admission.
Example dataset for using the tool
The example dataset is a sequence alignment file that is a smaller subset of the data generated only from ST258 isolates. This is provided in the inst/extdata directory of the package as example.FASTA. The epidemiological data matching these isolates is provided in the example dataset as example.RData. Dates have been shifted relative to the start date to protect patient privacy, but the structure of the data is the same as the complete dataset. The example dataset is used in the package documentation and vignettes to demonstrate how to use the package.
This example dataset is not too small, but small enough to still be manageable for testing and demonstration purposes. It is a subset of the complete dataset, which is much larger and contains more sequence types. The example dataset is used to demonstrate the functionality of the package without overwhelming users with too much data. Locally, when tested on an M1 MacBook Air, the example dataset only takes about a minute to run the clustering algorithm and generate the phylogenetic tree, which is a reasonable amount of time for a demonstration.