Short Manual for the FUNAGE-Pro Web Server


Input


Select Genome

In this section the reference genome is selected. If a hit will be shown in the blue bar below the input fields. NOTE: RefSeq and Genbank locus-tags are mostly not similar.
  • RefSeq: The reference sequence will be used from the Reference Sequence (RefSeq) Databases of NCBI
  • GenBank:The reference sequence will be used from the Genbank Databases of NCBI
  • FACoP: The sessionID provided by FACoP

Data table

Data should be provided as a tab-delimited table, e.g., copy and paste from Excel. Use Load Example Data to see the input formats.
VALUES: Data should consist of log transformed ratio (log2(ratio)) or Fold values. Both are accepted by FUNAGE-Pro.
  • Single list: A list of locus_tags, e.g., a set of over-expressed genes.
  • Single list with values:contains 2 columns; the first column contain the locus-tags and the second column the values. The expected values are Ratio values derived for a Differential Gene Expression (DGE) analysis.
  • Experiments: Multiple experiments such as time series, can be analyzed in one go. The first column should contain the locus_tags and all other columns contain all the experiments.
  • Clusters: sClusters or Modules are derived from clustering analysis (e.g. k-means) or module discovery tools (e.g. from gene network reconstruction) Multiple columns are allowed here; The first column should contain the locus-tags and the column with the header clusterID is used for calling the clusters or modules

Auto detect threshold values

FUNAGE-Pro analysis can estimate the threshold values by benchmarking the data on the basis of GO terms.
  • ON: Threshold values will be estimated by FUNAGE-Pro.
  • OFF:Threshold values are provided by the user.

Analyse data as

FUNAGE-Pro analysis can estimate the threshold values by benchmarking the data on the basis of GO terms.
  • Single list: for data with or without values. The first or 2 columns will be used, even if you entered more.
  • Experiments:for multiple experiments. We do not recommend to use this if you have one experiment only.
  • Clusters/Modules:the column with the header clusterID will be used to group the genes.


RESULTS

Summary Table

The nine classes (COG, GO, KEGG, KEYWORDS, PFAM, SMART, Superfamily and Operons) and the number of overrepresented ClassIDs (e.g. GO:0009082, GO:0005215, etc) of each experiment/cluster are shown in the summary table. A classID is overrepresented if it contains significant more genes than is expected on random basis. In principle, higher number of classIDs means more significant biological difference.

Main Table

The sortable Main Table contain the results of all experiments / clusters for all classes.
Select "Score, Hits/Class Size, p-value, or Gene set" before Exporting the table data.

The columns contain the following information:
  • ClassID: The class term and a link to term details. Note; KEYWORDs redirect to a Wiki page, but not all keywords are described in a Wiki pages
  • Class: The Class name, e.g., KEGG
  • Description: Description of the class term derived from the original class annotation
  • LC: Line Chart (LC) is a simple representation of the original input values
  • Heatmap: Similar data as the Line Chart but presented as a Heatmap
  • Experiments: Each experiment is presented in a column with 4 values which can be turned on or off enhancing readabilty.
    • Score: The score is a ranked from 0 to 9, poor to good, light to dark blue, respectivily.
    • Hits/Class Size:Number of genes found in the class / size of the class
    • p-value: p-value
    • Gene set: Gene IDs found in the class


Bar graphs

In case of multiple experiments (or clusters) analysis; for each of the nine classes 3 Bar Graphs are available. In case of a single experiment analysis, only Bar Graph 2 will be shown. Hovering over the bars will show the genes (aka hits) of the selected ClassID.

  • Bar Graph 1: Number of overrepresented ClassIDs (Functional Classes) per Experiment (or Cluster). Clicking or hovering over the bars will show all ClassIDs of the selected bar in Bar Graph 2.
  • Bar Graph 2: The significance of all ClassIDs of the selected bar from Bar Graph 1. The p-value is presented in the minus log scale, meaning that higher bars represent more signifcant overrepresentation of the ClassID. Clicking or hovering over the bars will activate Bar Graph 3.
  • Bar Graph 3: The selected ClassID from Bar Graph 2 is show for all experiments (or Clusters). The bar size represents the number of significant genes (aka Hits).

Export data: Each Bar graph has a download button for "Image" and "Data". Before exporting the image, the Label Size, Label Angle and Title Size van be changed using the sliders.