README.md

In sgrote/NanoStringNormalizeR: Normalize all NanoString .RCC files in a directory

NanoStringNormalizeR

Normalize all NanoString .RCC files in a directory using NanoStringNorm

Installation

if (!requireNamespace("devtools", quietly=TRUE))
    install.packages("devtools")
devtools::install_github("sgrote/NanoStringNormalizeR")

Example workflow

1. Load this package

library(NanoStringNormalizeR)

2. Set a path to the directory that contains the .RCC files

For example:

input_folder = "C:/Users/user/NanoString/Raw_Data"

3. Define housekeeping genes

For example:

house_genes = c("ACTB", "GUSB", "MRPL19", "PSMC4", "PUM1", "RPLP0", "SF3A1", "TFRC")

4. Normalize all .RCC files

This puts all the .RCC files in input_folder together and normalizes them at once:

norm_fc(input_folder, house_genes)

Note that also .RCC_ files in subdirectories of input_folder are taken into account._

This created a folder results in the input_folder, e.g.

C:/Users/user/NanoString/Raw_Data/results/

containing the following files:

file | description | ----- | ----- | raw_data.csv | unnormalized NanoString data | normalized_data.csv | normalized NanoString data | flagged_samples.txt | flagged samples (unusual normalization factors) | ratio.csv | ratios for target samples | ratio_mvp_reference.csv | ratios for reference samples (mvp-samples) | fold_change.csv | fold-changes for target samples | fold_change_mvp_reference.csv | fold-changes for reference samples (mvp-samples) | ratio_without_mvp_reference.csv | ratio for target samples compared to all target samples | fold_change_without_mvp_reference.csv | fold-changes for target samples compared to all target samples |

Note that existing files will be overwritten.

5. Plot ratios of normalized data

To create a .png file with a heatmap of ratios, set a path to a .csv file with ratios. This can be the ratio.csv created above, e.g.

ratio_csv = "C:/Users/user/NanoString/Raw_Data/results/ratio.csv"

While this could be plotted right away, ratio.csv will often contain too many data for a good looking plot. In those cases it is useful to filter it ratio.csv first, e.g. with a list of genes. To filter for genes, set a path to a .csv file with the genes you want to keep in the first column, e.g.

genes_csv = "C:/Users/user/NanoString/pam50_genes.csv"

Then filter ratio_csv with it:

subset_ratios(ratio_csv, genes_csv, "ratio_pam50_subset.csv")

This will create a file ratio_pam50_subset.csv in the same folder as ratio_csv.

To plot this subset as a heatmap run

ratio_subset = "C:/Users/user/NanoString/Raw_Data/results/ratio_pam50_subset.csv"
plot_ratios(ratio_subset)

This will create a file ratio_heatmap.png in the same folder as ratio_pam50_subset.csv.

To see more options for the heatmaps run

?plot_ratios

Of course the subset of ratio.csv could also have been filtered with any other tool, e.g. in Excel. Also, the subsetting and plotting would work in the same way with fold_change.csv as input.

sgrote/NanoStringNormalizeR documentation built on Dec. 30, 2020, 7:45 p.m.