New features include speed improvements to runGSA() and a new GSA method option: fgsea.
Now, gsea is ~100x faster than before. The fgsea method is ~30x faster than the default mean method, and completes within seconds.
Piano 1.10 is now available!
There is a bug fix regarding the reported number of up and down regulated genes for the GSEA method of runGSA.
Piano 1.8.0 is now available!
This is a minor update conforming to updated R requirements.
Check out our new tool Kiwi for GSA result visualization and interpretation!
Piano 1.6.0 is now available!
New features include:
- A function writeFilesForKiwi enabling visualization of GSA results with our new software tool Kiwi
- Minor bug fixes
(see Change log for details)
Piano 1.4.0 is now available!
New features include:
- Improved control and customization of the consensusHeatmap
- New function GSAheatmap for single GSA runs
- Several minor improvements, bugfixes and documentation updates
(see Change log for details)
Uncomfortable with the R command line? Check out BioMet-Toolbox.org where you can run some of the most important functions of piano in a point-and-click based browser interface!
Important microarray analysis bug-fix in piano 1.2.5. Please update!
The bug was in diffExp() and in some rare cases returned incorrect gene IDs for $pValues and
$foldChanges in the returned result object. This happened if and only if annotation was previously supplied to loadMAdata()
AND the probeset IDs in the normalized data ($dataNorm) were unsorted. The information in
$resTable as well as in the saved tables (if using save=TRUE) was always correct and unnaffected.
Piano 1.2.0 is now available!
New features include:
- runGSA runs up to 90% faster
- the new runGSAhyper function for hypergeometric test
- new plots and more control over existing plots
- several minor improvements, bugfixes and documentation updates
(see Change log for details)
The devel version of piano (version 1.1.15) includes some new features: volcano plots are now part of the output of diffExp, the gene names in the Venn diagram can now be accessed, a new function runGSAhyper has been added for peforming a hypergeometric test. See below how to use the latest devel version of piano.
The devel version of piano (version 1.1.13) includes a major improvement in runtime of runGSA, which now can run up to 90% faster using multiple CPUs, controlled through the new argument ncpus. See below how to use the latest devel version of piano.
Piano is now released and hosted on Bioconductor.
Piano is an R-package (www.r-project.org) and part of Bioconductor (www.bioconductor.org). Piano is used for running gene set analysis with various statistical methods, from different gene level statistics and a wide range of gene-set collections. Furthermore, the piano package contains functions for combining the results of multiple runs of gene set analyses.
The piano package consists of two parts:
1. The major part revolves around gene set analysis (GSA), and the central function for this is runGSA. There are some downstream functions (e.g. GSAsummaryTable and geneSetSummary that handle the results from the GSA. By running runGSA multiple times with different settings it is possible to compute consensus gene set scores. Another set of functions (e.g. consensusScores and consensusHeatmap take a list of result objects given by runGSA for this step.
2. The second part of the piano package contains a set of functions devoted for an easy-to-use approach on microarray analysis (wrapped around the affy and limma packages), which are constructed to integrate nicely with the downstream GSA part. The starting function in this case is loadMAdata.
Email the developer Leif Väremo at piano.rpkg[.at.]gmail.com
If you encountered any problems please check if your questions are already answered on the help page, in the vignette or in the function documentation before you email for support. Thank you!
This work was carried out at Systems and Synthetic Biology at Chalmers University of Technology, Gothenburg, Sweden.
Some of the functionalities of piano are available through the browser-based GUI BioMet Toolbox. Here, you can upload your data and apply a selection of the most important functions and options that you want to run, using a point-and-click based system.
Advanced: Using devel version
For advanced users who want to access the latest development updates of piano it is possible to install the devel version. There are at least three alternatives:
Alt 1. Simply run: > install.packages("piano",repos="http://bioconductor.org/packages/devel/bioc/", type="source")
Alt 2. You can also download the appropriate file (piano_x.y.z.tar.gz, piano_x.y.z.zip or piano_x.y.z.tgz) from here, and install manually, e.g. install.packages('piano_x.y.z.tar.gz', repos=NULL, type="source").
Alt 3. It is possible to run the whole Bioconductor as devel version, in which case packages can be installed as usual, see this page for more info.
Code development is available through GitHub.
Bioconductor also hosts GitHub mirrors of the release and devel versions of all its packages.
Q: I have a question about piano, how do I get support? A: First, check if your question in answered on this help page. If not, read the documentation
for the specific function you are using (e.g. run in R: > ?runGSA to read about runGSA).
A lot of the functionality of piano is described with examples in the vignette (see Documentation above). If none of the above helped you, feel free to contact the developers for assistance (see next question).
Q: How can I get in contact with the developers? A: Email the developer Leif Väremo at piano.rpkg[.at.]gmail.com
If you encountered any problems please check if your questions are already answered on this page, in the vignette or in the function documentation before you email for support. Thank you!
Q: I have problems installing piano, what is wrong? A: First be sure to update to the latest version of R. In general it is difficult to say what the problem is. Read the error and warning messages carefully. Is there any mentions of other packages that failed to install. In that case, try to install them separately first.
Q: The function runGSA() is taking a long time, can I speed it up? A: Yes! You can run this function on multiple CPUs using the argument ncpus. Be sure to install the package snowfall first. Reducing the number of permutations using the argument nPerm will also make it faster, but also decrease the resolution (unique number) of the gene set p-values. The number of gene sets can be limited by only considering ones within a given size range (number of genes) using the argument gsSizeLim, resulting in a faster run.
Please cite this publication when you use piano:
Väremo, L., Nielsen, J. and Nookaew, I. (2013) Enriching the gene set analysis of genome-wide data by incorporating directionality of gene expression and combining statistical hypotheses and methods. Nucleic Acids Research. 41 (8), 4378-4391. DOI:10.1093/nar/gkt111 [link]
Please cite this publication if you use the BioMet Toolbox 2.0:
Garcia-Albornoz, M., Thankaswamy-Kosalai, S., Nilsson, A., Väremo, L., Nookaew, I. and Nielsen, J. (2014) BioMet Toolbox 2.0: genome-wide analysis of metabolism and omics data. Nucleic Acids Research. 42 (W1), W175-W181. DOI:10.1093/nar/gku371 [link]
Please cite this publication if you use Kiwi:
Väremo, L., Gatto, F. and Nielsen, J. (2014) Kiwi: a tool for integration and visualization of network topology and gene-set analysis. BMC Bioinformatics. 15 (408). DOI:10.1186/s12859-014-0408-9 [link]
A selection of publications using piano:
A full list of citations is available e.g. in Google Scholar
Mülleder, M., Calvani, E., Alam, M.T., Wang, R.K., Eckerstorfer, F., Zelezniak, A. and Ralser, M. (2016)
Functional Metabolomics Describes the Yeast Biosynthetic Regulome. Cell DOI:10.1016/j.cell.2016.09.007 [link]
Monk, J.M., Koza, A., Campodonico, M.A., Machado, D., Seoane, J.M., Palsson, B.O., et al. (2016)
Multi-omics Quantification of Species Variation of Escherichia coli Links Molecular Features with Strain Phenotypes. Cell Systems DOI:10.1016/j.cels.2016.08.013 [link]
Caspeta, L., Chen, Y., Ghiaci, P. et al. (2014) Altered sterol composition renders yeast thermotolerant. Science 346:75. DOI:10.1126/science.1258137 [link]
Gatto, F., Nookaew, I. and Nielsen, J. (2014) Chromosome 3p loss of heterozygosity is associated with a unique metabolic network in clear cell renal carcinoma. PNAS DOI:10.1073/pnas.1319196111 [link]
Gatto, F., Schulze, A. and Nielsen, J. (2016)
Systematic Analysis Reveals that Cancer Mutations Converge on Deregulated Metabolism of Arachidonate and Xenobiotics. Cell Reports 16:3. DOI:10.1016/j.celrep.2016.06.038 [link]
Väremo, L., Scheele, C., Broholm, C., Mardinoglu, A., et al. (2015) Proteome- and transcriptome-driven reconstruction of the human myocyte metabolic network and its use for identification of markers for diabetes. Cell Reports. 11:6. DOI:10.1016/j.celrep.2015.04.010 [link]
Rudolph, K.L.M., Schmitt, B.M., Villar, D., White, R.J., Marioni, J.C., Kutter, C. and Odom, D.T. (2016) Codon-Driven Translational Efficiency Is Stable across Diverse Mammalian Cell States. PLOS Genetics. DOI:10.1371/journal.pgen.1006024 [link]
Tyakht, A.V., Kostryukova, E.S., Popenko, A.S., et al. (2013) Human gut microbiota community structures in urban and rural populations in Russia. Nature Communications. 4:2469. DOI:10.1038/ncomms3469 [link]
Mardinoglu, A., Agren, R., Kampf, C. et al. (2014) Genome-scale metabolic modelling of hepatocytes reveals serine deficiency in patients with non-alcoholic fatty liver disease. Nature Communications. 5:3083 DOI:10.1038/ncomms4083 [link]
Shoaie, S., Karlsson, F., Mardinoglu, A., et al. (2013) Understanding the interactions between bacteria in the human gut through metabolic modeling. Scientific Reports. 3:2532. DOI:10.1038/srep02532 [link]
Kildegaard, K.R, Hallström, B.M, Blicher, T.H. et al. (2014) Evolution reveals a glutathione-dependent mechanism of 3-hydroxypropionic acid tolerance. Metabolic Engineering 26 DOI:10.1016/j.ymben.2014.09.004 [link]
Nam, D. (2015) Effect of the absolute statistic on gene-sampling gene-set analysis methods. Statistical Methods in Medical Research. DOI:10.1177/0962280215574014 [link]