This study highlights the limitations of common data transformations used to fit transcriptomic data, particularly single-cell RNAseq data, into a normal distribution. Through simulation, it demonstrates that such transformations can distort graphical structures and affect the interpretation of results in two-sample problems, emphasizing the need for specialized methods tailored for count data.
Banzato, E., Risso, D., Chiogna, M., Djordjilović, V. (2025). Data Transformation and Its Validity in a Two-Sample Problem: An Illustration Based on Graphical Models. Springer Nature [10.1007/978-3-031-64431-3_21].
Data Transformation and Its Validity in a Two-Sample Problem: An Illustration Based on Graphical Models
Erika Banzato
;Monica Chiogna;
2025
Abstract
This study highlights the limitations of common data transformations used to fit transcriptomic data, particularly single-cell RNAseq data, into a normal distribution. Through simulation, it demonstrates that such transformations can distort graphical structures and affect the interpretation of results in two-sample problems, emphasizing the need for specialized methods tailored for count data.File in questo prodotto:
Eventuali allegati, non sono esposti
I documenti in IRIS sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione.
