From ae6575bcbe2a84991bb8b5eb6dccc2ad9ba9197d Mon Sep 17 00:00:00 2001 From: JacobsonMT Date: Wed, 21 Mar 2018 14:57:07 -0700 Subject: [PATCH] Resolve #36: Help on stability is incomplete. Adds more contextual help in enrichment view, specifically to each column and row of row expansion. Reorganizes the overlaypanel to be closer to their calling button. Related to issue #37. --- gotrack/src/main/webapp/enrichment.xhtml | 210 ++++++------------ .../composites/enrichmentResultsTable.xhtml | 192 ++++++++++++++-- .../src/main/webapp/resources/css/common.css | 12 +- .../main/webapp/resources/css/enrichment.css | 4 +- 4 files changed, 254 insertions(+), 164 deletions(-) diff --git a/gotrack/src/main/webapp/enrichment.xhtml b/gotrack/src/main/webapp/enrichment.xhtml index 5480628..f6d403d 100644 --- a/gotrack/src/main/webapp/enrichment.xhtml +++ b/gotrack/src/main/webapp/enrichment.xhtml @@ -62,6 +62,9 @@ + +

Select the species you would like to run an enrichment analysis for.

+ @@ -116,6 +119,9 @@ + +

Which aspects (Cellular Component, Biological Process, Molecular Function) to include in the analysis.

+ @@ -132,6 +138,13 @@ + +

Type of multiple tests correction to apply.

+
    +
  • Bonferroni: Reject null hypotheses who P-values are less than the given threshold.
    • +
  • BH step-up: Benjamini–Hochberg procedure, controls FDR at given threshold level.
    • +
+ @@ -147,6 +160,10 @@ + +

Threshold to use for the above multiple test correction method. This establishes the definition of "significant" for the results displays.

+ + @@ -162,6 +179,9 @@ + +

Select only those terms with at least this number of genes associated with them. Used to to remove overly-specific terms from analysis.

+ @@ -176,6 +196,11 @@ + + +

Select only those terms with at most this number of genes associated with them (0 for no maximum). Used to remove overly-general terms from analysis.

+ + @@ -195,6 +220,10 @@ + +

The analysis will generate a similarity plot. This plot can compare subsequent editions to each-other (PROXIMAL) or compare each edition to the most current (CURRENT). +

+ @@ -254,12 +283,46 @@
+ +
+

Description:

+

GO Term Counts by Edition shows counts of unique GO Terms associated with this each enrichment analysis over time.

+

Legend Descriptions:

+

All Tested Terms: Count of unique terms that passed all settings criteria.

+

Significant Terms: Count of unique terms that passed all settings criteria and were significant.

+

Rejected Terms: Count of unique terms that did not pass all settings criteria.

+ +

Controls:

+

<Click> any edition to view the enrichment results at that point in time in bottom table.

+

<Click> a legend item to toggle that series.

+

<Ctrl/Command> + <Click> a legend item to show only that series.

+
+
+ +
+

Description:

+

Enrichment Similarity shows Jaccard similarity of the sets of significantly enriched terms over time.

+ + +

Legend Descriptions:

+

The series differ in which sets of terms we are comparing:

+

All Terms: Compare all significant terms.

+

Top 5 Terms: Compare the top 5 terms.

+

Parents of Top Terms: Compare the top 5 terms and all of their ancestors.

+

Genes Backing Top Terms: Compare the genes that are responsible for the top 5 terms.

+ +

Controls:

+

<Click> any edition to view the term sets described above.

+

<Click> a legend item to toggle that series.

+

<Ctrl/Command> + <Click> a legend item to show only that series.

+
+ @@ -268,6 +331,17 @@
+ +
+

These buttons provide a means to visualize the changes in enrichment results over the + history of this hitlist. You may choose from plotting either the top 5 terms in every + edition or your selected terms By Rank or By P-Value.

+

If By Rank is chosen, you will be shown a Bump Chart of the corresponding terms + and their relative ranks over time. Areas of heavy change will have lots of crossing + lines (and thus appear darker) while areas of stability will have lots of parallel + lines (and thus appear lighter and cleaner)

+
+
- - -

Select the species you would like to run an enrichment analysis for.

- - -

Which aspects (Cellular Component, Biological Process, Molecular Function) to include in the analysis.

- - -

Type of multiple tests correction to apply.

-
    -
  • Bonferroni: Reject null hypotheses who P-values are less than the given threshold.
    • -
  • BH step-up: Benjamini–Hochberg procedure, controls FDR at given threshold level.
    • -
- - - - -

Threshold to use for the above multiple test correction method. This establishes the definition of "significant" for the results displays.

- - - - -

Select only those terms with at least this number of genes associated with them. Used to to remove overly-specific terms from analysis.

- - -

Select only those terms with at most this number of genes associated with them (0 for no maximum). Used to remove overly-general terms from analysis.

- - -

The analysis will generate a similarity plot. This plot can compare subsequent editions to each-other (PROXIMAL) or compare each edition to the most current (CURRENT). -

- - - -
-

Description:

-

GO Term Counts by Edition shows counts of unique GO Terms associated with this each enrichment analysis over time.

-

Legend Descriptions:

-

All Tested Terms: Count of unique terms that passed all settings criteria.

-

Significant Terms: Count of unique terms that passed all settings criteria and were significant.

-

Rejected Terms: Count of unique terms that did not pass all settings criteria.

- -

Controls:

-

<Click> any edition to view the enrichment results at that point in time in bottom table.

-

<Click> a legend item to toggle that series.

-

<Ctrl/Command> + <Click> a legend item to show only that series.

-
- - - -
-

Description:

-

Enrichment Similarity shows Jaccard similarity of the sets of significantly enriched terms over time.

- - -

Legend Descriptions:

-

The series differ in which sets of terms we are comparing:

-

All Terms: Compare all significant terms.

-

Top 5 Terms: Compare the top 5 terms.

-

Parents of Top Terms: Compare the top 5 terms and all of their ancestors.

-

Genes Backing Top Terms: Compare the genes that are responsible for the top 5 terms.

- -

Controls:

-

<Click> any edition to view the term sets described above.

-

<Click> a legend item to toggle that series.

-

<Ctrl/Command> + <Click> a legend item to show only that series.

-
- - - - - -
-

Description:

-

Enrichment Results Table shows enrichment data for those terms that were significantly enriched at some point in their history.

-

- This means that not every term in table is significant at the currently chosen time-point, but it was necessarily significant as some point. -

- - -

Column Descriptions:

-

: View more details, such as which genes in your hit list are associated with the term.

-

Id: Gene Ontology Id. <Click> on the icons for links to QuickGO and its ancestry chart, respectively.

-

Aspect: One of Biological Process, Cellular Component, or Molecular Function.

-

Name: Description of the Gene Ontology Term.

-

Set Size: Number of genes in the background (currently all annotated genes) that are annotated with the given term.

-

Hits: Number of genes in the hitlist annotated with the given term.

-

P-Value: P-Value (after correction if Bonferroni).

-

- Recent Stability: Metric representing the variability of the term's p-value in its recent past (akin to the logarithm of the coefficient of variation). This is calculated using - six month rolling regressions on the historical changes in 'Set Size' and 'Hits'. <Click> on - to plot the term's p-value and 95% confidence interval over its history. -

-

- Note: The stability score typically ranges between -10 and 10 with 10 being more stable than -10. A score of ∞ means we saw no significant variation in Set Size or Hits over a given timeline (keep in mind this does - not mean the p-value is constant). A score with � or NaN means there was not enough available data. -

- -
- - - - -
-

Description:

-

These buttons provide a means to visualize the changes in enrichment results over the history of this hitlist. You may choose from plotting either the top 5 terms in every edition or your selected terms By Rank or By P-Value.

-

If By Rank is chosen, you will be shown a Bump Chart of the corresponding terms and their relative ranks over time. Areas of heavy change will have lots of crossing lines (and thus appear darker) while areas of stability will have lots of parallel lines (and thus appear lighter and cleaner)

- -
- - \ No newline at end of file diff --git a/gotrack/src/main/webapp/resources/composites/enrichmentResultsTable.xhtml b/gotrack/src/main/webapp/resources/composites/enrichmentResultsTable.xhtml index be3f707..83b2404 100644 --- a/gotrack/src/main/webapp/resources/composites/enrichmentResultsTable.xhtml +++ b/gotrack/src/main/webapp/resources/composites/enrichmentResultsTable.xhtml @@ -38,6 +38,18 @@ oncomplete="postAjaxSortTable(PF('tableEnrichmentWdg'))"/> + +
+

This table shows enrichment data for those terms that were significantly enriched in this edition as well as those that were significant at some point in their history.

+

+ This means that not every term in the table is significant at the currently chosen edition, but it was necessarily significant as some point. +

+
+ + @@ -72,11 +84,45 @@ + + +
+

<Click> on to view more details, such as which genes in your hit list are associated with the term.

+ +
+ + + + filterMatchMode="contains" + sortBy="#{entry.term.goId}" + style="width:120px;" + exportable="false"> + + +
+

+ Gene Ontology Id. +

+

+ <Click> on to view the term's QuickGO entry. +

+

+ <Click> on to view the term's ancestry chart, NOTE: we do not propagate across aspects whereas QuickGo does. +

+
+ + + + @@ -92,9 +138,24 @@ - + + +
+

+ One of Biological Process, Cellular Component, or Molecular Function. +

+
+ + + + + filterMatchMode="contains" + sortBy="#{entry.term.name}"> + + +
+

+ Short description of the given term. +

+
+ + + + + style="width:80px;"> + + +
+

+ Number of genes in the background (all annotated genes) that are annotated with the given term. +

+
+ + + + + + + + +
+

+ Number of genes in the hit list annotated with the given term. +

+
+ + + + - + + +
+

+ P-Value (after correction if Bonferroni). Highlighted green if the term is significant in the selected edition. +

+
+ + + + @@ -144,11 +262,44 @@ - + + +
+

Metric representing the variability of the term's p-value in its recent past.

+

+ This is calculated using six month rolling regressions on the historical changes in + 'Pop. Hits' and 'Hits'. Given the results of these regressions we use their 95% confidence + intervals to calculate a 95% confidence interval around the p-value. This range is then used + to calculate the negative log of the coefficient of variation with respect to that edition's + significance cutoff to get the stability score. +

+

+ The stability score typically ranges between -10 and 10 with 10 being more + stable than -10. +

+

+ A score of No Change means we saw no significant variation in both + Pop. Hits and Hits over its recent past (keep in mind this does not mean the p-value + is constant). +

+

A score of No Data means there was not enough available.

+

+ <Click> on to plot the term's + p-value and 95% confidence interval over its history. +

+
+ + + + @@ -166,53 +317,54 @@ - + + - + - + - + - + - + - + - + - + diff --git a/gotrack/src/main/webapp/resources/css/common.css b/gotrack/src/main/webapp/resources/css/common.css index 07656cc..50a1ce8 100644 --- a/gotrack/src/main/webapp/resources/css/common.css +++ b/gotrack/src/main/webapp/resources/css/common.css @@ -95,8 +95,8 @@ button.ui-button.icon-only span { button.ui-button.no-value { height: 12px; - width: 2.2em; - outline:none; + width: 1.5em; + outline: none; } button.ui-button.no-background { @@ -114,6 +114,8 @@ button.ui-button.no-background:hover { button.ui-button.no-value > span.ui-icon { margin-left: -8px; margin-top: -8px; + height: 100%; + width: 100%; } button.ui-button.no-value > span.ui-button-text { @@ -264,9 +266,9 @@ div.inline { padding: 5px; } -.overlay-help { - padding: 0.5em 0; -} +/*.overlay-help {*/ + /*padding: 0.5em 0;*/ +/*}*/ .overlay-help p { margin: 0.2em 1em; diff --git a/gotrack/src/main/webapp/resources/css/enrichment.css b/gotrack/src/main/webapp/resources/css/enrichment.css index fe039b5..fd433b6 100644 --- a/gotrack/src/main/webapp/resources/css/enrichment.css +++ b/gotrack/src/main/webapp/resources/css/enrichment.css @@ -215,7 +215,7 @@ tr.ui-widget-content.significant td:nth-last-child(2) { } div.ui-overlaypanel-content { - width: 600px; + max-width: 600px; text-align: left; } @@ -266,7 +266,5 @@ button.ui-button.absoluteBtn { } button.ui-button.absoluteBtn > span { - left: 5px; - top: 5px; z-index: 99999; } \ No newline at end of file