POMA EDA Example

Pol Castellano-Escuder1*

1Duke University

*polcaes@gmail.com

Contents

Compiled date: 2023-04-25

Last edited: 2022-08-02

License: GPL-3

1 Installation

Run the following code to install the Bioconductor version of package.

# install.packages("BiocManager")
BiocManager::install("POMA")

2 Load POMA

library(POMA)

3 Automatic EDA Report

The following function will return an Exploratory Data Analysis (EDA) HTML report. The input object must be a SummarizedExperiment object.

data("st000336")
PomaEDA(st000336)

Generated EDA HTML report starts here.

4 Know your data

• Your data have 57 samples, 31 features and 2 groups, that are ‘Controls’, ‘DMD’. Furthermore, 1 covariates have been found in your data. These covariates are ‘steroids’.

• A 3.45% of values in your data are NAs (missing values). Variables that have NA values are ‘aspartic_acid (5)’, ‘citrulline (28)’, ‘glutamic_acid (15)’, ‘methionine (3)’, ‘x3_hba (1)’, ‘a_kg (1)’, ‘citrate (2)’, ‘fumarate (1)’, ‘lactate (2)’, ‘malate (1)’, ‘pyruvate (1)’, ‘succinate (1)’.

• A 0% of values in your data are zeros.

• Removed from the exploratory data analysis 0 features that only have zeros.

• Removed from the exploratory data analysis 0 features that have zero variance.

4.1 Summary Tables

  Samples Features Covariates
1      57       31          1
  Number_Zeros Percentage_Zeros
1            0              0 %
  Number_Missings Percentage_Missings
1              61              3.45 %

4.2 Samples by Group

5 Normalization Plots

6 Group Distribution Plots

7 Outlier Detection

7 possible outliers detected in your data. These outliers are ‘DMD119.2.U02’, ‘DMD084.11.U02’, ‘DMD087.12.U02’, ‘DMD023.10.U02’, ‘DMD046.11.U02’, ‘DMD133.9.U02’, ‘DMD135.10.U02’.

# A tibble: 7 × 4
  sample        group    distance_to_centroid limit_distance
  <chr>         <fct>                   <dbl>          <dbl>
1 DMD119.2.U02  Controls                 2.74           2.29
2 DMD084.11.U02 DMD                      4.52           4.04
3 DMD087.12.U02 DMD                      4.25           4.04
4 DMD023.10.U02 DMD                      5.65           4.04
5 DMD046.11.U02 DMD                      5.58           4.04
6 DMD133.9.U02  DMD                      5.35           4.04
7 DMD135.10.U02 DMD                      4.06           4.04

8 High Correlated Features (r > 0.97)

There are 0 high correlated feature pairs in your data.

9 Heatmap and Clustering

10 Principal Component Analysis

11 Uniform Manifold Approximation and Projection Clustering