IPA spring release 2016

Published Date: April 13th, 2016: IPA┬« is a powerful web-based knowledge base system designed for biologists.

Read the full release notes (here).

Content Updates

~95,000 new findings (bringing total to nearly 5.5 million findings), including:

  • ~54,000 new Expert findings
  • ~700 new ExpertAssist findings
  • ~28,000 new mutation-to-disease findings from ClinVar
  • ~5,000 new cancer mutation disease association findings from COSMIC
  • ~2,300 new protein-protein interactions from the BioGRID database
  • ~2,000 new disease-to-drug findings from ClinicalTrials.gov
  • ~1,400 new disease target findings from ClinicalTrials.gov
  • ~2,200 new gene-to-biological process assertions from Gene Ontology (GO)
  • ~600 new mouse knockout-to-phenotype findings in Mouse Genome Database (JAX Labs)
  • ~600 new protein-protein interactions from the IntAct database ~160 gene-to-disease findings from OMIM

Support for additional microarray:

  • Agilent chip: SurePrint G3 Human GE v3 8x60K Microarray

Mapping and Gene Models

Identifier source versions:

  • Affymetrix: na35 (01/18/2016)
  • Agilent: various (as of 01/18/2016)

  • DBSNP: b146 (bovine, dog, human, mouse, rat)
  • b145 (chicken)
  • b142 (zebrafish)
  • b138 (Arabidopsis, nematode)
  • b137 (drosophila)
  • (http://www.ncbi.nlm.nih.gov/projects/SNP/snp_summary.cgi)

  • EntrezGene: 01/18/2016

  • Ensembl: Ensembl 83 (12/2016) (http://uswest.ensembl.org/info/website/archives/index.html)

  • Genbank: NCBI-GenBank Flat File Release 211 (12/18/2015)
  • (ftp://ftp.ncbi.nih.gov/genbank/release.notes/)

  • HGNC: 01/18/2016

  • HomoloGene: HomoloGene Release 68 (04/09/2014)
  • (http://www.ncbi.nlm.nih.gov/homologene/statistics/)

  • Illumina: 10/25/2010

  • MirBase: version 21 (June 2014)
  • RefSeq: Release 72 (11/05/2015)
  • (ftp://ftp.ncbi.nlm.nih.gov/refseq/release/release-notes/archive/)

  • UCSC18hg: 3/2006
  • UCSC19hg: GRCh37 (2/2009)

  • Unigene: Human(#236), Mouse(#194), Rat(#195), Bovine (#100), Chicken (#46), Dog (#25), Fruitfly (#75), Nematode (#52), Zebrafish(#126), Arabidopsis(#74) (ftp://ftp.ncbi.nih.gov/repository/UniGene/, log files)

  • Uniprot: UniProt release 2015_12, 12/09/2015
  • (ftp://ftp.uniprot.org/pub/databases/uniprot/previous_releases/)

Gene model source versions:

  • Hg38/mm38:
  • Ensembl Human: Homo_sapiens.GRCh38.83.gtf.gz
  • Ensembl Mouse: Mus_musculus.GRCm38.83.gtf.gz

  • RefSeq human: GCF_000001405.32_GRCh38.p6
  • RefSeq mouse: GCF_000001635.24_GRCm38.p4

  • Hg19/mm10 from UCSC:
  • Ensembl Human: April 7th 2014
  • Ensembl Mouse: April 7th 2014

  • RefSeq Human: June 30th 2014
  • RefSeq Mouse: June 30th 2014

WhatÔÇÖs New in the IPA Spring Release (March 2016)

Quickly compare results across ÔÇÿomics datasets on networks and pathways

Identify significant trends in genes involved in a pathway or network across conditions such as time or dose and elucidate possible mechanisms driving gene expression results with both variant gain or loss of function and expression results. Visualize multiple ÔÇÿomics datasets simultaneously on IPA networks and pathways.

Overlay multiple gene expression datasets/analyses on a canonical pathway (or on any collection of genes) simultaneously to see how genes are regulated across various conditions. Visualize multiple measurements at onceÔÇöfor example both Fold Change and the Intensity of the expression (e.g. RPKM in the case of RNA-seq data) as shown in Figure 1.

(see picture on the original page)

IPA also presents the multi-dataset / multi-measurement results in a table view that can be exported.

(see picture on the original page)

Elucidate possible mechanisms driving gene expression results by simultaneously overlaying both gene expression analysis and variant loss/gain datasets on a pathway or network. In this way you can see which genes are differentially expressed and harbor potentially deleterious variants.

(see picture on the original page)

The new overlay is useful even when showing a single dataset or analysis The table and bar charts enable you to see multiple measurements for each gene at once.The Matching Molecules table shows the ÔÇ£Display nameÔÇØ for genes on the pathway which sometime differ from the genesÔÇÖ official gene symbols. For example, in the pathway image shown below, the gene FERMT2 (gene symbol) from the dataset is displayed as KIND2 (common name) on the pathway. The Matching Molecules table makes it easy to make the connection between the gene symbol and common names often used in publications.Frequently a gene from the dataset will be part of a group (gene family) or protein complex on a Canonical Pathway with a name that differs from the individual gene symbol. Again the Matching Molecule table enables you to easily determine to which node(s) it belongs to on the pathway.Tip: You can quickly find the corresponding gene in the pathway by selecting it in the Matching Molecules table which outlines the matching node(s) in dark blue on the pathway image to the right of the overlay panel (not shown below)

(see picture on the original page)

Other Application Improvements

Easier-to-understand metabolic pathways. Now itÔÇÖs easy to see which enzymes or enzyme families catalyze each step in IPAÔÇÖs metabolic pathways; reactions are now labeled with enzyme names rather than E.C. numbers.

(see picture on the original page)

Enhance your pathway exploration by selecting & expanding multiple groups at onceCanonical Pathway diagrams often contain nodes that represent groups (gene families). Now you can expand more than one node at a time to display which genes or proteins they contain. Just select a subset (or all) of the nodes on a pathway and right-click and select ÔÇ£Show members/ MembershipÔÇØ to expand to reveal their members. The added nodes are automatically laid out in a circular ÔÇ£cloudÔÇØ pattern in one area on the pathway. Hint: sometimes it helps to expand small sets of nodes at a time to control the way the member nodes are laid out.

(see picture on the original page)

Quickly focus on upstream regulators of interest by filtering in Upstream Regulator AnalysisNow you can filter the Upstream Regulator Analysis results table by the name of the regulator of interest to display only those rows of importance to your area of study. You may use the asterisk as a wildcard to find all that match.

(see picture on the original page)

Focus your Core Analyses by reducing the types of molecules that are usedIn some cases, it may be desirable to eliminate certain type(s) of molecules from participation in your Core Analysis, especially in Causal Networks, Mechanistic Networks, or Interaction Networks. For example, since you cannot directly measure the expression of groups (gene families) or protein complexes in gene expression experiments, you may want to avoid including them in the networks that are generated in Core Analysis. Therefore, Core Analysis now provides a filter ÔÇ£accordionÔÇØ option to select node types to include or exclude from the analysis.

(see picture on the original page)

Important note: Keep in mind that selections that you make in this new accordion limit the node types that IPA adds into networks and that are scored from your dataset against pathways and in functional analysis. If you exclude kinases for instance, then the kinases that were differentially expressed in your dataset will not be allowed to participate in the analysis, and will not be scored against canonical pathways, diseases and functions, and will not be included IPAÔÇÖs networks.

Improved control over the color scale for networks and pathway overlay in Comparison Analysis. Now you can manually enter a range for the color scale for a Gene Heatmap in Comparison Analysis. For example, you can type in -10 to 10 for expression fold change, which will mean that the green color will saturate (e.g. be at its brightest) at genes with -10 fold or lower fold change, and the red will saturate at genes with +10 or higher fold change. Furthermore, the corresponding color range on the pathway or network shown in the right panel in Comparison Analysis now adjusts to match the range settings.

(see picture on the original page)

Enhance performance of large Comparison Analyses with higher memory settings. With this release, if you are running IPA on a 64-bit computer with ample installed and available RAM, we recommend you set IPA to use 4000, 6000, or even 8000 Mb of RAM in the IPA preferences panel if you work with large Comparison Analyses. This will improve performance when opening and navigating large analyses, especially if comparing many analyses that contain a significant number of causal networks.

Note: Regardless of how you use IPA we recommend that you set IPA to use at least 1000 Mb (rather than the default 750 Mb) if you have enough available RAM to do so.


To access IPA at VIB, please check our IPA page


To update IPA core analyses to the new content, read our IPA-FAQ page (VIB users only).