NOJAH : NOt Just Another Heatmap

This interactive web application: NOt Just Another Heatmap (NOJAH) is developed in R with Shiny to

1) Perform Genome-Wide Heatmap (GWH) Analysis on any cancer genomic data set
2) Perform Combined results Clustering (CrC) Analysis for up to three different data types.
3) Perform Significance of Cluster (SoC) Analysis using a robust bootstrap approach

The goal of this tool is to provide a one stop shop to perform genomic analyses.

Input GW Data

Select an example dataset or upload your own with 'Load my own GW data.'

Choose file to upload (maximum size 50 MB)

Browse...

Download GW TCGA BRCA Expression DataSet

Genome-Wide Dendrogram Display

Plot GW Dendrogram (May take up to a few minutes to load)

Yes

Data Subsetting

Subset GW data by:

Variance

Median Absoute Deviation

Inter Quartile Range

Percentile

Percentile Slider

Manually Enter Percentile

Percentile Value:

Percentile value

Percentile

Percentile Slider

Manually Enter Percentile

Percentile Value:

Percentile value

Percentile

Percentile Slider

Manually Enter Percentile

Percentile Value:

Percentile value

Percentile

Percentile Slider

Manually Enter Percentile

Percentile Value:

Percentile value

Percentile

Percentile Slider

Manually Enter Percentile

Percentile Value:

Percentile value

Percentile

Percentile Slider

Manually Enter Percentile

Percentile Value:

Percentile value

Percentile

Percentile Slider

Manually Enter Percentile

Percentile Value:

Percentile value

Hit button to update results after each change in input parameter(s)

Download subset data

Type the file name you would like to save subset data as :

Download Subset data

Subset HeatMap Input

Select an pre-existing dataset or upload your own with 'Load my own subset data.'

Choose file to upload (maximum size 50 MB)

Browse...

Download subset Pre-existing DS

Hit button to update results after each change in input parameter(s)

Input Settings

Select first row and column where numeric data starts

Data row

Data column

Download subset Heatmap

Save Heatmap as :

Download HM

Download Column clusters after cut-tree

Download Row clusters after cut-tree

Consensus Clustering Input

Select an pre-existing dataset or upload your own with 'Load my own GW data.'

Choose file to upload (maximum size 50 MB)

Browse...

Download subset Pre-existing DS

Hit button to update results after each change in input parameter(s)

Download Consensus Cluster Results

Download may take a while. Once complete, the result pdf file will automatically open.

Silhouette Input

Select an pre-existing dataset or upload your own with 'Load my own subset data.'

Choose file to upload (maximum size 50 MB)

Browse...

Download subset Pre-existing silhouette DS

Select an pre-existing cluster or upload your own with 'Load my own clusters.'

Choose file to upload (maximum size 50 MB)

Browse...

Download subset Pre-existing silhouette clusters

Hit button to update results after each change in input parameter(s)

Download Silhouette Core Samples

Input Core Data

Select an example dataset or upload your own with 'Load my own Core data.'

Choose file to upload (maximum size 50 MB)

Browse...

Hit button to update results after each change in input parameter(s)

Input Settings

Select first row and column where numeric data starts

Data row

Data column

Core Sample based Downloads

Type the file name you would like to save subset data as :

Download Subset data

Save HeatMap as :

Download HM

Download Column clusters after cut-tree

Download Row clusters after cut-tree

A Genome-Wide Heatmap can be very dense. Given the limitation with the computational power required to construct a genome wide heatmap, NOJAH showcases a Genome-Wide Dendrogram.

Genome-Wide Heatmap Analysis workflow is divided into four main subparts:

1. Define Core Features with Most Variable Approach

2. Heatmap of Core Features

3. Define Cluster Number

4. Define Core Samples

Heatmap is updated based on the Core Features with Core Samples.

When using the analysis workflow, each step of the workflow is intended to be used sequentially i.e. the output of step 1 is fed into step 2 as input and so on. However each of these components can also be used independently.For example, if only consensus clustering needs to be performed then the 'Cluster Number' tab can be used.

Core Features
Heatmap
Cluster Number
Core Samples
Updated Heatmap
Workflow

Genome-Wide Dendrogram

Measures of Spread

Define Core Features with Most Variable Approach

To see the position of your 'gene of interest', enter gene id in the text box to the right

Heatmap of Core Features

Heatmap
Column Dendrogram
Row Dendrogram

Define Number of Clusters

Define Core Samples

Heatmap of Core Features with Core Samples

Heatmap
Column Dendrogram
Row Dendrogram

Workflow for Genome-Wide Heatmap (GWH) Analysis

Genome-Wide Dendrogram Options

Normalization Type

Normalize by:

Scale Range

Distance Method

Agglomerative Linkage Method

Dendrogram Label size:

Enter the 'gene id' you wish to annotate. Please note that input is case sensitive. Entering only gene name instead of gene id will throw error. For example, when using preloaded data enter: SCGB1D2>ENSG00000124935.3

Heat Map Options

Normalization Type

Normalize by:

Scale Range

Plot Margin dimensions

Clustering Measures

Distance Method

Agglomerative Linkage Method

Row dendrogram

TRUE FALSE

Col dendrogram

TRUE FALSE

Display Row labels?:

No Yes

If yes, Row Label font size

Display Col labels?:

No Yes

If yes, Col Label font size

Heat Map colors

low

mid

high

Adjust Column Label font size

Adjust Row Label font size

Cut Col dendrogram?:

No Yes

Cut Col Dendrogram at:

Cut Row dendrogram?:

No Yes

Cut Row Dendrogram at:

Consensus Clustering Options

Distance Measure

Clustering Method

Maximum number of clusters:

No. of iterations

100

500

1000

Choose Optimal Number of clusters

Optimal k is

Silhouette Options

Distance Measure

Remove samples based on:

Fixed value

Percentile

Change point

Remove Samples within Cluster1 with Silhoutte width less than:

Remove Samples within Cluster2 with Silhoutte width less than:

Remove Samples within Cluster3 with Silhoutte width less than:

Remove Samples within Cluster4 with Silhoutte width less than:

Remove Samples within Cluster5 with Silhoutte width less than:

Remove Samples within Cluster6 with Silhoutte width less than:

Remove Samples within Cluster7 with Silhoutte width less than:

Remove Samples within Cluster8 with Silhoutte width less than:

Remove Samples within Cluster9 with Silhoutte width less than:

Remove Samples within Cluster1 with percentile silhoutte width less than:

Remove Samples within Cluster2 with percentile silhoutte width less than:

Remove Samples within Cluster3 with percentile silhoutte width less than:

Remove Samples within Cluster4 with percentile silhoutte width less than:

Remove Samples within Cluster5 with percentile silhoutte width less than:

Remove Samples within Cluster6 with percentile silhoutte width less than:

Remove Samples within Cluster7 with percentile silhoutte width less than:

Remove Samples within Cluster8 with percentile silhoutte width less than:

Remove Samples within Cluster9 with percentile silhoutte width less than:

Changes Using:

Method:

Max Q Allowed:

Remove Samples within Cluster1 with silhoutte width greater than change point:

Remove Samples within Cluster2 with silhoutte width greater than change point:

Remove Samples within Cluster3 with silhoutte width greater than change point:

Remove Samples within Cluster4 with Silhoutte width greater than change point:

Remove Samples within Cluster5 with Silhoutte width greater than change point:

Remove Samples within Cluster6 with Silhoutte width greater than change point:

Remove Samples within Cluster7 with Silhoutte width greater than change point:

Remove Samples within Cluster8 with Silhoutte width greater than change point:

Remove Samples within Cluster9 with Silhoutte width greater than change point:

Heat Map Options

Normalization Type

Normalize by:

Scale Range

Plot Margin dimensions

Clustering Measures

Distance Method

Agglomerative Linkage Method

Row dendrogram

TRUE FALSE

Col dendrogram

TRUE FALSE

Display Row labels?:

No Yes

If yes, Row Label font size

Display Col labels?:

No Yes

If yes, Col Label font size

Heatmap colors

low

mid

high

Adjust Column Label font size

Adjust Row Label font size

Cut Col dendrogram?:

No Yes

Cut Col Dendrogram at:

Cut Row dendrogram?:

No Yes

Cut Row Dendrogram at:

Input file

Select the example Expression ds or upload your own with 'Load my own'

Choose file to upload (maximum size 500 MB).

Browse...

Download TCGA BRCA Expression data

Select the example Meth/Variant ds or upload your own with 'Load my own'

Choose file to upload (maximum size 500 MB).

Browse...

Download TCGA BRCA Meth data

Select the example CNV ds or upload your own with 'Load my own'

Choose file to upload (maximum size 500 MB).

Browse...

Download TCGA BRCA CNV data

Select the example CrC analysis ds or upload your own with 'Load my own'

Choose file to upload (maximum size 500 MB).

Browse...

Download CrC analysis data

Data type

Choose data used

Methylation

Variant

Clustering Options

Distance Measure

Clustering Method

No. of iterations

100

500

1000

Hit button to update results after each change in input parameter(s)

Distance Measure

Clustering Method

No. of iterations

100

500

1000

Hit button to update results after each change in input parameter(s)

Distance Measure

Clustering Method

No. of iterations

100

500

1000

Hit button to update results after each change in input parameter(s)

Perform CrC analysis using

Expression

Methylation/Variant

Copy Number

Select atleast two platforms to run CrC analysis

Distance Measure

Clustering Method

No. of iterations

100

500

1000

Hit button to update results after each change in input parameter(s)

Clinical markers

Select a clinical feature from below or add your own using 'Load your own'.

Download TCGA BRCA clinical ds Example

Add clinical features (Yes/No)

Number of features

Choose file to upload (maximum size 500 MB).

Browse...

Download Results

Consensus Clustering

Download Expression clusters

Download may take a while. Once complete, the result pdf file will automatically open.

Download Meth/Variant clusters

Download may take a while. Once complete, the result pdf file will automatically open.

Download CNV clusters

Download may take a while. Once complete, the result pdf file will automatically open.

Download CrC analysis

Download may take a while. Once complete, the result pdf file will automatically open.

Expresssion
Methylation/Variant
Copy Number
Cluster of Clusters (CoC) Analysis

Choose Optimal Number of clusters

Optimal k is

Clustering Measures

Distance Method

Agglomerative Linkage Method

Row dendrogram

TRUE FALSE

Col dendrogram

TRUE FALSE

Display Row labels?:

No Yes

If yes, Row Label font size

Display Col labels?:

No Yes

If yes, Col Label font size

Heat Map Options

Plot Margin dimensions

Contingency Table(s) options:

Starify by:

Expression

Methylation/Variant

Copy Number

Input Data to test significance of clusters

Select an example dataset or upload your own with 'Load my own GW data.'

Choose file to upload (maximum size 50 MB)

Browse...

Download Example DataSet

Input Settings

Select first row and column where numeric data starts

Data row

Data column

Hit button to update results after each change in input parameter(s)

Type the file name you would like to save as

Download HeatMap and dendrograms

Download Column clusters after cut-tree

Download Row clusters after cut-tree

HeatMap
Column Dendrogram
Row Dendrogram

Heat Map Options

Normalization Type

Normalize by:

Scale Range

Plot Margin dimensions

Clustering Measures

Distance Method

Agglomerative Linkage Method

Row dendrogram

TRUE FALSE

Col dendrogram

TRUE FALSE

Display Row labels?:

No Yes

If yes, Row Label font size

Display Col labels?:

No Yes

If yes, Col Label font size

Heat Map colors

low

mid

high

Adjust Column Label font size

Adjust Row Label font size

Cut Col dendrogram?:

No Yes

Cut Col Dendrogram at:

Assess Gene set significance in separation of specimens into 2 clusters?:

No Yes

Select a dataset or upload your own with 'Load my own data.'

Choose file to upload to sample from to estimate significance of separation (Maximum size 100 MB)

Browse...

Gene-Set size:

No. of bootstrap samples:

Click the button to start sampling using bootstrap method for estimating the p-value. A progress indicator will appear shortly (~approx 10 seconds), on top of page indicating the status. Once complete, the p-value will be displayed in the main panel.

Cut Row dendrogram?:

No Yes

Cut Row Dendrogram at:

Assess significance of samples in separation of gene set into 2 clusters (only for more than 2 cluster groups)?:

No Yes

Select a dataset or upload your own with 'Load my own data.'

Choose file to upload to sample from to estimate significance of separation (Maximum size 75 MB)

Browse...

Sample-set size:

No. of bootstrap samples:

The National Cancer Institute (NCI) requires that publications acknowledge the Winship Cancer Institute CCSG support, and they are tracking compliance. When using this tool to report results in your publication, please include the following statement in the acknowledgment section of your publication(s):

Research reported in this publication was supported in part by the Biostatistics and Bioinformatics Shared Resource of Winship Cancer Institute of Emory University and NIH/NCI under award number P30CA138292. The content is solely the responsibility of the authors and does not necessarily represent the official views of the National Institutes of Health.

Authors- Manali Rupji, dual M.S., Bhakti Dwivedi Ph.D. & Jeanne Kowalski Ph.D.
Maintainer- Manali Rupji 'manali(dot)rupji(at)emory(dot)edu'

The Biostatistics and Bioinformatics Shared Resource at Winship Cancer Institute of Emory University

https://bbisr.winship.emory.edu/

This App is developed and maintained by Manali Rupji at the Biostatistics and Bioinformatics core, Winship Cancer Institute, Emory University.

As a Biostatistics and Bioinformatics core, we are actively improving and expanding our NGS analysis services and analysis products. For any questions, comments, or suggestions, please email the developer at mrupji@emory.edu.

Input GW Data

Genome-Wide Dendrogram Display

Data Subsetting

Hit button to update results after each change in input parameter(s)

Download subset data

Subset HeatMap Input

Hit button to update results after each change in input parameter(s)

Input Settings

Select first row and column where numeric data starts

Download subset Heatmap

Download Column clusters after cut-tree

Download Row clusters after cut-tree

Consensus Clustering Input

Hit button to update results after each change in input parameter(s)

Download Consensus Cluster Results

Download may take a while. Once complete, the result pdf file will automatically open.

Silhouette Input

Hit button to update results after each change in input parameter(s)

Download Silhouette Core Samples

Input Core Data

Hit button to update results after each change in input parameter(s)

Input Settings

Select first row and column where numeric data starts

Core Sample based Downloads

Download Column clusters after cut-tree

Download Row clusters after cut-tree

A Genome-Wide Heatmap can be very dense. Given the limitation with the computational power required to construct a genome wide heatmap, NOJAH showcases a Genome-Wide Dendrogram.

Genome-Wide Heatmap Analysis workflow is divided into four main subparts:

Genome-Wide Dendrogram

Measures of Spread

Define Core Features with Most Variable Approach

To see the position of your 'gene of interest', enter gene id in the text box to the right

Heatmap of Core Features

Define Number of Clusters

Define Core Samples

Heatmap of Core Features with Core Samples

Workflow for Genome-Wide Heatmap (GWH) Analysis

Genome-Wide Dendrogram Options

Heat Map Options

Clustering Measures

Heat Map colors

Consensus Clustering Options

Choose Optimal Number of clusters

Silhouette Options

Heat Map Options

Clustering Measures

Heatmap colors

Input file

Data type

Clustering Options

Hit button to update results after each change in input parameter(s)

Hit button to update results after each change in input parameter(s)

Hit button to update results after each change in input parameter(s)

Select atleast two platforms to run CrC analysis

Hit button to update results after each change in input parameter(s)

Clinical markers

Download Results

Consensus Clustering

Download may take a while. Once complete, the result pdf file will automatically open.

Download may take a while. Once complete, the result pdf file will automatically open.

Download may take a while. Once complete, the result pdf file will automatically open.

Download may take a while. Once complete, the result pdf file will automatically open.

Sample Clusters

CoC HeatMap

Cluster Interpretation

Choose Optimal Number of clusters

Clustering Measures

Heat Map Options

Contingency Table(s) options:

Input Data to test significance of clusters

Input Settings

Select first row and column where numeric data starts

Hit button to update results after each change in input parameter(s)

Download Column clusters after cut-tree

Download Row clusters after cut-tree

Heat Map Options

Clustering Measures

Heat Map colors