Skip to contents

Multi-domain chronic wound skin microbiome data

Source: R/data.R
skin.Rd

The multi-domain skin microbiome dataset analysed in the paper, from the chronic wound study of Verbanic et al. (2020, 2022). Wound swabs were collected from 20 patients attending an outpatient wound care clinic: from the chronic wound before and after a debridement treatment, and from a control site of healthy skin, giving 60 samples from 20 subjects under three experimental conditions.

Format

A list with elements:

Y

A list of two integer count matrices, bacteria (60 x 75) and virus (60 x 39), samples in rows.

X

A 60 x 3 indicator matrix for the experimental condition, with columns pre, post and healthy, so that \(\beta_{mj1}\), \(\beta_{mj2}\) and \(\beta_{mj3}\) are the pre-treatment, post-treatment and healthy effects respectively.

subject

A length-60 integer vector giving the subject of each sample; three consecutive samples share a subject.

J

The number of OTUs per domain, c(75, 39).

Source

Verbanic, S., Shen, Y., Lee, J., Deacon, J. M. and Chen, I. A. (2020). Microbial predictors of healing and short-term effect of debridement on the microbiome of chronic wounds. npj Biofilms and Microbiomes 6(1), 21.

Verbanic, S., Deacon, J. M. and Chen, I. A. (2022). The chronic wound phageome: phage diversity and associations with wounds and healing outcomes. Microbiology Spectrum 10(3), e0277721.

Details

Bacterial abundance was measured by sequencing the V1-V3 loops of the 16S rRNA gene and aggregated at the genus level (bOTUs); viral abundance was measured by sequencing DNA from virus-like particles and aggregated at the host level (vOTUs). Only OTUs with a non-zero count in at least two samples under each condition and an average count above ten under each condition were retained, leaving 75 bOTUs and 39 vOTUs. 42.98% of the bacterial and 44.10% of the viral counts are zero.

References

Zhang, S., Shen, Y., Chen, I. A. and Lee, J. (2025). Sparse Bayesian Group Factor Model for Feature Interactions in Multiple Count Tables Data. Journal of the American Statistical Association, 120(550), 723–736. doi:10.1080/01621459.2025.2449721

See also

spbgfm

Examples

data(skin)
vapply(skin$Y, dim, integer(2))
#>      bacteria virus
#> [1,]       60    60
#> [2,]       75    39
colSums(skin$X)
#>     pre    post healthy 
#>      20      20      20