Package 'classifyBLCA'

Centroids for the BLCA Consensus Classes

Description

A data.frame containing the centroid coordinates for the six consensus classes of muscle-invasive bladder cancer

Usage

centroids_consensus

Format

An object of class tbl_df (inherits from tbl, data.frame) with 857 rows and 9 columns.

Author(s)

Aurelie Kamoun

References

https://www.biorxiv.org/content/10.1101/488460v2

---

Centroids for the BLCA TCGA Classes

Description

A data.frame containing the centroid coordinates for the five TCGA classes of muscle-invasive bladder cancer

Usage

centroids_tcga

Format

An object of class tbl_df (inherits from tbl, data.frame) with 1264 rows and 7 columns.

Author(s)

Gordon Robertson

References

https://doi.org/10.1016/j.cell.2017.09.007

---

Centroids for the BLCA UROMOL2021 Classes

Description

A data.frame containing the centroid coordinates for the four UROMOL2021 classes of non-muscle-invasive bladder cancer

Usage

centroids_uromol2021

Format

An object of class tbl_df (inherits from tbl, data.frame) with 1942 rows and 7 columns.

Author(s)

Sia Lindskrog

References

https://www.nature.com/articles/s41467-021-22465-w

---

Calculate centroid correlation for samples

Description

Calculate centroid correlation for samples

Usage

classify_blca(
  df,
  gene_id = c("entrez", "ensembl", "hgnc"),
  classifier = c("tcga", "uromol2021", "consensus"),
  tidy = FALSE
)

Arguments

df	A data.frame containing RNA expression with unique genes as rows and samples as columns. RNA-seq data must be log-transformed.
gene_id	Character specifying the type of gene identifiers used for the row names or first column of df.
classifier	Classifier to be used.
tidy	Logical. If TRUE, assumes the first column contains the gene identifiers. Otherwise, assumes IDs are row names

Details

If using the TCGA classifier, you must provide either Entrez IDs or HGNC symbols. The original centroids provided for the TCGA classifier only included Entrez and HGNC, and the conversion from Entrez to Ensembl is ambiguous. Therefore, conversion and selection of unique genes must be performed by the user on a case-by-case basis, perhaps using the expression value of the gene as a guide for which genes to keep.

Value

A tibble containing:

estimate

Pearson correlation of a given sample to the given class centroid

conf.low, conf.high

low and high end of 95% confidence interval

nearest

centroid for which the sample has the highest correlation to

statistic

t-statistic

parameter

degrees of freedom

sep_lvl

(Highest Correlation - 2nd Highest Correlation)/median(Distance to highest correlation)

Author(s)

Aurelie Kamoun

---