LipidSig 2.0: Integrating Lipid Characteristic Insights into Advanced Lipidomics Data Analysis

In Differential Expression Page, significant lipid species or lipid characteristics can be explored through two main customised analysis, by ‘Lipid species’ or by ‘Lipid characteristics’ , with user-uploaded data. Subsequently, further analysis and visualisation methods, including dimensionality reduction, hierarchical clustering, characteristics analysis, and two characteristic analysis, can be implemented based on the results of differential expressed analysis by utilising user-defined methods and characteristics.

Demo dataset source (two-group): Adipose tissue ATGL modifies the cardiac lipidome in pressure-overload-induced left Ventricular failure (PLoS Genet. 2018)
Demo dataset source (multi-group): Lipidomic and biophysical homeostasis of mammalian membranes counteracts dietary lipid perturbations to maintain cellular fitness (Nat Commun 11, 1339. 2020)

Data Source

Data processing

Remove features with many missing values

More than % missing values

Fill missing value with:

Multiply by minimum

Tune sigma

nPCs

The number of neighbors

nPcs

Data Normalization

Normalization with:

Data Transformation

Transformation with:

Processed data

Download

Lipid species analysis
Lipid characteristics analysis

Differential expression

In lipid species analysis section, differentially expressed analysis is performed to find significant lipid species. In short, samples will be divided into two/multiple groups (independent) based on the Group Information of input. Two statistical methods are provided for different data types: the t-test and the Wilcoxon test (Wilcoxon rank-sum test)for two-group data, along with one-way ANOVA and the Kruskal-Wallis test for multi-group data. Additionally, the p-value will be adjusted using the Benjamini-Hochberg procedure. The condition and cut-offs for significant lipid species are also users selected.

Method:

Post Hoc method:

Tukey's HSD

Multiple testing correction:

Benjamini & Hochberg

Identify significant lipids:

p-value:

Fold change (FC):

Result download

The results of the Differential Expression Analysis are formatted as the input directly for the Enrichment Analysis. Click the button below to download the data, then proceed to the Enrichment webpage to upload it.

Download Enrichment table

The results from the Differential Expression Analysis are formatted to serve directly as input for the Network Analysis. Once you download the data using the button below, you will find folders named by network analysis type, each corresponding to a different network analysis. Proceed to the appropriate Network Analysis web page section and upload the input data from the relevant folder.

Download Network table

Dimensionality reduction

Dimensionality reduction is common when dealing with large numbers of observations and/or large numbers of variables in lipids analysis. It transforms data from a high-dimensional space into a low-dimensional space so that to retain vital properties of the original data and close to its intrinsic dimension.

Dimensionality reduction method:

Scaling:

Centering:

NOTE: Scaling and Centering are on by default. Centering is strongly recommended for pre-processing steps.

Scaling:

PCA:

Perplexity:

Number of iterations:

Scaling:

Number of neighbors:

Distance metric:

Clustering method:

Number of groups:

Distance metrics:

Number of groups:

Distance measure:

Clustering method:

Epsilon:

minPts:

Hierarchical clustering

Lipid species that derived from two/multiple groups will be clustered and visualised on heatmap using hierarchical clustering. Through heatmap, users may discover the difference between the two/multiple groups by observing the distribution of lipid species. This analysis provides an overview of lipid species differences between the control group and the experimental group.

Select feature for clustering:

Side color by lipid characteristics:

Distance measure:

Clustering method:

Characteristics association

In this part, we categorize significant lipid species based on different lipid characteristics and visualise the difference between control and experimental groups by applying log2 Fold Change.

Result table of differiential expression analysis

Lollipop chart of significant lipid species

The lollipop chart presents lipid species that meet the predefined cut-off criteria.
The x-axis indicates the log2 fold change for two-group data and the -log10 p-value for multi-group data. Lipid species are listed along the y-axis. The color of each point on the chart corresponds to the -log10 adjusted p-value or p-value.

Hover the mouse on the plot to view the corresponding detailed information.

For figure manipulation, please refer to FAQ.

For two-group data, the MA plot categorizes lipid species into three groups: up-regulated (red), down-regulated (blue), and non-significant (grey). Similarly, the volcano plot follows the same color scheme to highlight the most biologically significant lipid species visually.

For multiple-group data, the scatter plot of significant expressed lipid species in each class. The dot color is corresponding to the lipid class.

Hover the mouse on a specific dot (lipid species) to view the corresponding detailed information.

Hover the mouse on a specific dot (lipid species) to display the boxplot on the right-hand side with their abundance by groups.

For figure manipulation, please refer to FAQ.

This section displays the abundance boxplot and table of a lipid species. Use the drop-down menu to select a significantly expressed lipid species. Once a specific lipid species is chosen, its abundance boxplot and corresponding table will be displayed.

Select a lipid:

Lipid species abundance boxplot

Download

Differential expression analysis result (lipid species)

The PCA plot visually simplifies and discerns patterns within complex lipidomic data, effectively reducing multidimensional variables to principal components. The distinct separation or overlap of the groups reflects the underlying differences or similarities.

The scree plot is a common method for determining the number of PCs to be retained. The "elbow" of the graph indicates all components to the left of this point can explain most variability of the samples.

Hover the mouse on the plot to view the corresponding detailed information.

For figure manipulation, please refer to FAQ.

PCA plot

PCA scree plot

Rotation table

Table of PCA contribution table

The correlation circle plot illustrates the relationship between top N individual features (lipid species) and principal components (PCs). It displays how all the variables are interrelated, with those positively correlated positioned in the same quadrant and negatively correlated ones located diametrically across the origin of the plot.

The feature contribution histogram offers an in-depth view of how individual features (lipid species) contribute to a user-selected principal component, such as PC1, PC2, or a combination thereof (PC1+PC2). It allows users to identify which features influence the chosen principal component more.

Adjust the slider above each plot to choose the desired number of top features to display.

Hover the mouse on the plot to view the corresponding detailed information.

For figure manipulation, please refer to FAQ.

top N feature:

PCA correlation circle plot

top N feature:

Contribution of features to principal component:

Feature contribution histogram

The PLS-DA plot visually simplifies and discerns patterns within complex lipidomic data, effectively reducing multidimensional variables to principal components. The distinct separation or overlap of the groups reflects the underlying differences or similarities.

In the PLS-DA varialble loading plot, the distance to the variables' center indicates the variable's contribution. The value of the x-axis reveals the contribution of the variable to PLS-DA-1, whereas the value of the y-axis discloses the contribution of the variable to PLS-DA-2.

Hover the mouse on the plot to view the corresponding detailed information.

For figure manipulation, please refer to FAQ.

Table of sample variate

Table of sample loading

The t-SNE plot visually simplifies and discerns patterns within complex lipidomic data, effectively reducing multidimensional variables to principal components. The distinct separation or overlap of the groups reflects the underlying differences or similarities.

Hover the mouse on the plot to view the corresponding detailed information.

For figure manipulation, please refer to FAQ.

t-SNE plot

Table of t-SNE data

The UMAP plot visually simplifies and discerns patterns within complex lipidomic data, effectively reducing multidimensional variables to principal components. The distinct separation or overlap of the groups reflects the underlying differences or similarities.

Hover the mouse on the plot to view the corresponding detailed information.

For figure manipulation, please refer to FAQ.

UMAP plot

Table of UMAP data

The heatmap's top annotation categorizes data by sample group, while the side annotation—selectable from a drop-down menu—allows for organization according to various lipid characteristics. Cell colors reveal the correlation values, with red indicating positive correlation and blue signifying negative correlation.

If the lipid/sample number exceeds 50, their names will not be displayed on the heatmap.

Hover the mouse on the plot to view the corresponding detailed information.

For figure manipulation, please refer to FAQ.

Lipid species hierarchical clustering heatmap

Download matrix

The relevant results will appear upon choosing a lipid characteristic from the drop-down menu provided.

The lollipop chart compares all significant groups within the selected characteristic by log2(fold change) for two-group data and -log10(p-value) for multiple-group data.

The word cloud visualizes the frequency of each group (value) associated with the chosen characteristic.

The bar chart distinguishes significant groups (values) exhibiting a mean fold change greater than 2, with orange indicating significance and black denoting non-significance.

The bar chart is will not generate for multiple-group data.

Hover the mouse on the plot to view the corresponding detailed information.

For figure manipulation, please refer to FAQ.

Lollipop chart of all significant groups

Word cloud with the count of each group

Bar chart of significant groups

Differential expression

In Lipid Characteristics Analysis section, lipid species are categorised and summarised into new lipid abundance table according to two selected lipid characteristic, then conducted differential expressed analysis. Samples will be divided into two/multiple groups based on the Group Information of input data. Two-way ANOVA and appropriate post hoc tests for different data types are utilized. For two-group data, the t-test and the Wilcoxon test (Wilcoxon rank-sum test) are applied, while for multi-group data, one-way ANOVA and the Kruskal-Wallis test are used. This Differentially Expressed Analysis section separates into 2 sections, analysing based on first ‘Characteristics’ and adding ‘Subgroup of characteristics’ to the analysis. The first section is analysed based on the first selected ‘characteristics’ . The second section is the subgroup analysis of the first section.

Lipid characteristics table

Select a aspect:

NOTE: two selected characteristics should be both continuous data or one categorical data with one continuous data.

Method:

Two-way ANOVA

Identify significant lipids:

Post hoc tests:

p-value:

Multiple testing correction:

Benjamini & Hochberg

Fold change (FC):

Dimensionality reduction

Dimensionality reduction in this section assists users to tackle with large numbers of variables in lipids analysis. The high-dimensional space is transformed into a low-dimensional space. Hence, the crucial properties of the lipid data are revealed and still close to its intrinsic characteristics. Here, we provide four types of dimensionality reduction approaches, PCA, PLS-DA, t-SNE, UMAP, and four clustering methods, K-means, partitioning around medoids (PAM), Hierarchical clustering, and DBSCAN.

Dimensionality reduction method:

Scaling:

Centering:

NOTE: Scaling and Centering are on by default. Centering is strongly recommended for pre-processing steps.

Scaling:

PCA:

Perplexity:

Number of iterations:

Scaling:

Number of neighbors:

Distance metric:

Clustering method:

Number of groups:

Distance metrics:

Number of groups:

Distance measure:

Clustering method:

Epsilon:

minPts:

Hierarchical clustering

New lipid abundance table summed up from species will be clustered and shown on the heatmap using hierarchical clustering. Through heatmap, users may discover the difference between the two/multiple groups by observing the distribution of lipid characteristic abundance. This analysis provides an overview of lipid characteristic abundance differences between two/multiple groups. Four distance measures can be chosen, Person, Spearman, or Kendall, and eight clustering methods can be selected by pulling down the menu.

Select feature for clustering:

Distance measure:

Clustering method:

Two characteristics analysis

The heatmaps provide the correlation between the double bond and chain length of lipid species. The color in the heatmaps is gradient according to different data types (log2FC for two group data and p-value for multi-group data).

Method:

Identify significant lipids:

Post Hoc method:

Tukey's HSD

p-value:

Multiple testing correction:

Benjamini & Hochberg

Fold change (FC):

The plots displayed the differential expression analysis of a user-selected lipid characteristic.

The bar plot and line plot depict the different groups in each category of the lipid characteristic. All significant groups will be highlighted with an asterisk.

The box plot reveals each group's distribution of a selected lipid characteristic. Asterisks indicate the significance of P values: three for P values less than 0.001, two for less than 0.01, and one for less than 0.05. The absence of an asterisk or line denotes a non-significant difference between groups.

Switch over the tab to view the origin or sqrt-scaled bar plot and line plot.

Hover the mouse on the plot to view the corresponding detailed information.

For figure manipulation, please refer to FAQ.

Result table of lipid characteristic differiential expression analysis

Raw
Sqrt scale

Bar plot of characteristics analysis

Line plot of characteristics analysis

Box plot of characteristics analysis

Sqrt-scaled bar plot of characteristics analysis

Sqrt-scaled line plot of characteristics analysis

Box plot of characteristics analysis

Subgroup analysis of lipid characteristics

In Subgroup analysis of lipid characteristics, lipid species will be further split by the characteristic that user-chosen in the second pull-down menu then undergo the first section analysis. Two-way ANOVA is also applied with t-test as post hoc tests, and the cut-offs of differentially expressed lipids are inputted by users.

The drop-down menu lists all the categories within the selected subgroup characteristic. The corresponding result plots will displayed as a specific category is chosen.

Switch over the tab to view the origin or sqrt-scaled bar plot and line plot.

Hover the mouse on the plot to view the corresponding detailed information.

For figure manipulation, please refer to FAQ.

Select a category:

Result table of subgroup analysis

Raw
Sqrt scale

Bar plot of subgroup analysis

Line plot of subgroup analysis

Box plot of subgroup analysis

Sqrt-scaled bar plot of subgroup analysis

Sqrt-scaled line plot of subgroup analysis

Box plot of subgroup analysis

The PCA plot visually simplifies and discerns patterns within complex lipidomic data, effectively reducing multidimensional variables to principal components. The distinct separation or overlap of the groups reflects the underlying differences or similarities.

The scree plot is a common method for determining the number of PCs to be retained. The "elbow" of the graph indicates all components to the left of this point can explain most variability of the samples.

Hover the mouse on the plot to view the corresponding detailed information.

For figure manipulation, please refer to FAQ.

PCA plot

PCA scree plot

Rotation table

Table of PCA contribution table

The correlation circle plot illustrates the relationship between top N individual features (lipid characteristics) and principal components (PCs). It displays how all the variables are interrelated, with those positively correlated positioned in the same quadrant and negatively correlated ones located diametrically across the origin of the plot.

The feature contribution histogram offers an in-depth view of how individual features (lipid characteristics) contribute to a user-selected principal component, such as PC1, PC2, or a combination thereof (PC1+PC2). It allows users to identify which features influence the chosen principal component more.

Adjust the slider above each plot to choose the desired number of top features to display.

Hover the mouse on the plot to view the corresponding detailed information.

For figure manipulation, please refer to FAQ.

top N feature:

PCA correlation circle plot

top N feature:

Contribution of features to principal component:

Feature contribution histogram

The PLS-DA plot visually simplifies and discerns patterns within complex lipidomic data, effectively reducing multidimensional variables to principal components. The distinct separation or overlap of the groups reflects the underlying differences or similarities.

In the PLS-DA varialble loading plot, the distance to the variables' center indicates the variable's contribution. The value of the x-axis reveals the contribution of the variable to PLS-DA-1, whereas the value of the y-axis discloses the contribution of the variable to PLS-DA-2.

Hover the mouse on the plot to view the corresponding detailed information.

For figure manipulation, please refer to FAQ.

Table of sample variate

Table of sample loading

The t-SNE plot visually simplifies and discerns patterns within complex lipidomic data, effectively reducing multidimensional variables to principal components. The distinct separation or overlap of the groups reflects the underlying differences or similarities.

Hover the mouse on the plot to view the corresponding detailed information.

For figure manipulation, please refer to FAQ.

t-SNE plot

Table of t-SNE data

The UMAP plot visually simplifies and discerns patterns within complex lipidomic data, effectively reducing multidimensional variables to principal components. The distinct separation or overlap of the groups reflects the underlying differences or similarities.

Hover the mouse on the plot to view the corresponding detailed information.

For figure manipulation, please refer to FAQ.

UMAP plot

Table of UMAP data

Columns are all sample, and rows are the significant characteristic group (value) selected in the first `Characteristics` section from Step1. Cell colors reveal the correlation values, with red indicating positive correlation and blue signifying negative correlation.

If the lipid/sample number exceeds 50, their names will not be displayed on the heatmap.

Hover the mouse on the plot to view the corresponding detailed information.

For figure manipulation, please refer to FAQ.

Lipid characteristic hierarchical clustering heatmap

Download matrix

Total FA
Each FA

The heatmaps illustrate the correlation between lipid species' double bond counts and chain lengths based on a lipid characteristic selected in Step 1. The correlation is visually represented by cell colors—red indicates a positive correlation, while blue indicates a negative. Significant correlations are highlighted with an asterisk sign on the plot.

Switch over the tab to view the heatmap of total FA or each FA.

Two characteristics correlation heatmap

Download PDF

Table of two characteristics analysis

Choose a specific characteristic from the lipid characteristic selected in Step 1 via the drop-down menu to explore the correlation between the double bond counts and chain lengths. The correlation is visually represented by cell colors—red indicates a positive correlation, while blue indicates a negative. Significant correlations are highlighted with an asterisk sign on the plot.

Select the lipid characteristics:

Two characteristics correlation heatmap of selected category

Download PDF

Table of two characteristics analysis of selected category

In this section, the abundance box plot displays data for lipid species based on the lipid characteristic selected previously. You can investigate the relationship between double bond counts and chain lengths by choosing a particular lipid species from the drop-down menu. An asterisk sign indicates significant differences between groups. The absence of an asterisk or line denotes a non-significant difference between groups.

Select the lipid characteristics:

Abundance box plot of selected lipid species

Download PDF

Table of selected lipid species

The heatmaps illustrate the correlation between lipid species' double bond counts and chain lengths based on a lipid characteristic selected in Step 1. The correlation is visually represented by cell colors—red indicates a positive correlation, while blue indicates a negative. Significant correlations are highlighted with an asterisk sign on the plot.

Switch over the tab to view the heatmap of total FA or each FA.

Two characteristics correlation heatmap

Download PDF

Table of two characteristics analysis

Choose a specific characteristic from the lipid characteristic selected in Step 1 via the drop-down menu to explore the correlation between the double bond counts and chain lengths. The correlation is visually represented by cell colors—red indicates a positive correlation, while blue indicates a negative. Significant correlations are highlighted with an asterisk sign on the plot.

Select the lipid characteristics:

Two characteristics correlation heatmap of selected category

Download PDF

Table of two characteristics analysis of selected category

In this section, the abundance box plot displays data for lipid species based on the lipid characteristic selected previously. You can investigate the relationship between double bond counts and chain lengths by choosing a particular lipid species from the drop-down menu. An asterisk sign indicates significant differences between groups. The absence of an asterisk or line denotes a non-significant difference between groups.

Select the lipid characteristics:

Abundance box plot of selected lipid species

Download PDF