Please try the demo files in the sidebar (Demo File Sets).

Introduction

The 'LandScape' visualization is frequently utilized to provide a systematic illustration of integrative data from multiple layers of batch samples, which are always compared to each other on certain attributes, such as genes and biological pathways mutated in cancers. This online 'LandScape' visualization is designed as a fixed part (histogram and gene-panels) with additional panels (e.g., age, gender, and histology). To visualize data, upload a CSV file in the required format and use sidebar options to customize the display.

LandScape Data (CSV file)

The uploaded CSV file must match the required format as specified below.

• header
  The first line of the file should be a header that contains column names as keys.
• rows
  Each row in the file should contain data for a sample.
• samples (SampleID column)
  The first column lists sample names in each row, with the key SampleID in the header line. (mandatory)
• gene-panel (g_ columns)
  Add columns for genes that you want to display. (mandatory)
  • The header line keyword follows g_[GeneName] format, such as 'g_TP53' and 'g_PTEN'.
  • Typically, the content of each table cell for one gene and one sample lists the mutations types, such as 'Mis', 'InDel', and 'Gain'. Use - for non-mutation, and N/A for not-available.
  • A cell may contain multiple mutation types (or values) using semicolon as separator, e.g., Mis;Loss.
  • Each cell allows three unique values at maximum. For example, Mis;Splice_Site;Mis;Loss is OK, while Mis;Splice_Site;InDel;Mis;Gain is not allowed because it contains 4 distinctive mutation types.
  • All values appeared in g_ columns are summarized to calculate the sample frequency in each gene, which is shown by the horizontal histogram at the 'left-area' of gene-panel.
  • Generally, in cancer research, genes are always annotated with more information, such as their pathway, GO ontology, certain comments, and P-value or Q-value representing their mutated significance in batch samples. These information (if given) will be shown at the 'right-area' of gene-panel as tags, matrix or histogram.
    • for each pathway, add a row with key g_Pathway as the SampleID in the first column and specify the Pathway name (e.g., Adhesion) for each gene in other columns. Use N/A for not-available. (displayed as tags)
    • for each GO ontology, add a row with key g_GO as the SampleID in the first column and specify the GO name (e.g., Metabolic) for each gene in other columns. Use N/A for not-available. (displayed as tags)
    • for comments, add a row with key g_Comments_[name] in the SampleID column, where [name] should be the comments name, e.g., g_Comments_COSMIC. Values for comments must be boolean (Y or N), numeric values (e.g., 10 or 15.3), or N/A. (displayed as matrix)
    • for P-value, add a row with key g_P_value in the SampleID column and P-value (e.g., 0.001 or 3E-8) for each gene in other cells. Use 'N/A' for not-available. (displayed as histogram)
    • for Q-value, all should be the same as P-value except that the key should be g_Q_value. (displayed as histogram)
• histogram (ht columns)
  It is possible to add multiple stacked histograms that show attributes' distribution at the top of the visualization.
  • Column names for histograms should follow the ht_[AttributeName] format, such as ht_Missence and ht_Truncated.
  • Only numeric values (e.g., 15, 23.9 or 0.327) are allowed as cell content, and it keeps up to three decimals.
  • Since the whole CSV file as a complete matrix, please use - to fill the intersecting cells between 'histogram' columns and 'gene-information' rows (i.e., these mentioned above: pathway, GO, comments, P-value, and Q-value, if exist).
  • To display multiple histograms, use ht2_[attr] for the second one, ht3_[attr] for the third one, and so on. To allow flexibility, ht1_ and ht_ are both treated as the first one.
  • Basically, histogram is simply named with their NO., e.g., ht2_ corresponds to 'Histogram 2'. Optionally, users can assign a name to each histogram by adding a bracketed value after ht. For example, ht2(NewName)_ will be named as 'Histogram NewName' rather than 'Histogram 2'. The cooresponding section of this histogram in the sidebar will also use this new name.
  • Although it's not common, user can omit ht columns to avoid displaying any histogram in the figure.
• additional panels
  Add columns to show additional panels (more information on samples) at the bottom of the visualization. Each panel might contain several attributes belonging to one category, e.g., panel 'Individual-Info' contains attributes like age, gender, smoking.
  • The column name key should follow the [PanelName]_[AttributeName] format, e.g., Individual-Info_age and Individual-Info_gender.
    Note that please avoid using '_' in PanelNames. Do not worry about panel naming because the sidebar provides options to customize the displaying name of each panel, where users can choose their preferred name such as 'Individual_Info' or 'Individual Info'.
  • The cell content could be strings, numeric values, or N/A.
    • strings: It is always used for classifications, e.g., gender ('Female' and 'Male'). We accept maximum six classifications besides N/A. An attribute will be considered as a classifications type as long as strings is found in any cell in its column.
    • numeric values: It is always used for continuously distributed numeric attributes, e.g., age. The attribute will be shown in gradient color ranging from the minimum value to the maximum value. Considering that sometimes the numeric values may be used for classifications, e.g., tumor stages (1,2,3), OR, users just want to design groups corresponding to several value ranges, we provide related options in the sidebar to cover such usages. Note here that the classifications sill needs to subject to the maximum six requirements.
  • We have two reserved PanelName short names: mtif for 'Metadata' and pw for 'Pathway'. Of course, you can just use the full original name as you wish.
  • By default, all panels will be displayed from top to bottom on the page in the order in which they appear in the uploaded CSV file.

Display Interactions

There are four types of interactions: Highlights, Tooltips, External Link and Download.

• Highlights
  When the mouse moves on the LandScape figure, the column (sample) and row (gene or attribute) it points to will be highlighted.
• Tooltips
  Tooltips will show necessary information of object that the mouse points to.
  • histogram: SampleID, and value of the stacked area where the mouse points.
  • gene-panel
    • middle-area: SampleID, gene name, and relevant mutation types.
    • left-area: Gene name, and count of relevant (mutated) samples.
    • right-area: At gene-comments matrix, Gene name, comments name, and comments value. At P/Q-value histogram, Gene name, and P/Q-value.
  • other-panels: SampleID, attribute name and value.
• External Link
  
  • gene name: links to GeneCards search webpage.
  • pathway: links to kegg pathway search webpage.
  • GO: links to amigo search webpage.
• Download
  One SVG file will be generated when the 'Download' button is clicked. Two themes are supplied: the default theme with a dark background and the light theme with white background. To use the light theme, please click the 'Light Theme' button.

Sidebar Functions

The sidebar provides diverse options to fine-tune the display, such as manage files, reset size and color, group and reorder objects, and so on.

• Files
  • Manage Files: checklist of CSV files uploaded previously, delete or download the CSV files.
  • Upload: upload LandScape CSV file. Note that the duplicated file name will be alerted and given a random postfix.
  • Choose: choose files uploaded previously. Note that this function is ONLY available to registered users (each account has certain storage).
  • File Sets: NOT available to this page.
• General
  • Common: reset column width; choose color for 'N/A' (effective globally).
  • Samples: reorder samples with a series of selected conditions OR manually. We prepare a list of conditions (sorted or reversed sorted), such as SampleID (ASCII), Histogram values, and attributes in additional panels. Note that the well-known 'Bisect sorting' of mutated genes in cancer research is also provided.
  • Panels: reorder and rename additional panels.
• Histogram
  If there are multiple histograms, they will be shown in separate sections one by one in the sidebar.
  • Settings: hide this histogram; reset the maximum value of the Y-axis; reset the Y-axis label text; display SampleID along the X-axis.
  • Data: reorder the stacked groups; reset colors of the stacked groups.
• Gene panel
  • Main: reorder genes manually; reset the colors of mutation types; group gene comments; reset gradient colors and value ranges of gene comments.
  • Left: group mutation types; reset colors of mutation groups; reset maximum value of the axis; reset the axis label text.
  • Right: choose to show (P/Q-value, Pathway, GO, or None); reset maximum value of P/Q-value histogram axis; reset the threshold red line of P/Q-value histogram.
• Additional panel(s)
  If there are multiple additional panels, they will be shown in separate sections one by one in the sidebar.
  • Settings
    • for classifications attribute, reorder the categories in legend (left-side), and reset colors of each classification.
    • for numeric value attribute, reset gradient colors and value ranges, OR, enable groups displaying with customized value ranges.
  • Reorder: reorder attributes in this additional panel.

Manual version=1.3, written by Dr. JIA Wenlong and Mr. LI Hechen on 2020-03-11.