This project is maintained by Pablo Cingolani (email@example.com)
Acknowledgements and special thanks to (in order of appearance):
- Adrian Platts (McGill): tons of feedback, feature suggestions, test cases, etc.
- Doug Ruden (WSU): feedback, features request, debugging.
- Louis Letourneau (McGill / Genome Quebec): bug fixes, features suggestions, etc.
- Dave OConnor (UW-Madison): feature suggestions, HIV genome
- Denis Reshetov (Rogaev lab): VCF output support and debugging on GTF parsing.
- Louis Letourneau (McGill / Genome Quebec): Maven project
- Mark DePristo, Eric Banks & David Rozen (Broad): VCF parsing and VCF output.
- Jonathan Fresnedo (University of California, Davis): P.Persica genome support
- GATK: integratetion was performed by Broad's Genome Analysis Team: David Rozen, Eric Banks and Mark DePristo
- Arno Velds (NKI: Netherlands Cancer Institute): Annomalities in ENSEMBL annotation files that led to '-onlyCoding = auto' development, as well as "rogue transcript filters".
- Giulio Genovese: ENSEMBL annotation problems, fixing and improving "rogue transcript" problem.
- Louis Letourneau (McGill / Genome Quebec): dbNSFP implementation
- Peter Briggs (Manchester University) : Improved Galaxy interface and wrappers.
- Jim Johnson (Minnesota Supercomputing Institute, University of Minnesota): SnpSift improvements and bugfixes.
- Davide Cittaro (Center for Translational Genomics and Bioinformatics, Ospedale San Raffaele): Debugging and beta-testing for both SnpEff & SnpSift. Helped to debug many interoperability, compatibility and format specification issues.
- Louis Letourneau (McGill / Genome Quebec): Cancer effects, HSGV notation, etc.
- Sarmady, Mahdi (Children's hospital of Philadelphia): hg19 using transcript version from UCSC.
- Leipzig, Jeremy & Sarmady, Mahdi (Children's hospital of Philadelphia): HGSV notation
- Jinghua (Frank) Feng (University of Adelaide) : Losts of feedback, debugging and improvements in SnpSift annotations.
Please send any bug reports to:
Important: Try SnpEff's latest version. SnpEff is updated often, may be the bug has already been corrected (e.g you are using version 3.1A, but I've already fixed it in version 3.1B).
Note: A bug report should include at least
- An explanation of the bug condition
- A minimal bug-demonstrating test case (i.e. include SNP file used to run the test case)
- The conditions for the bug to occur (e.g. genome version)
- SnpEff version and sub-version numbers you are using (e.g. "I'm using SnpEff version 3.1 revision K")
I always try to help people, but sometimes I'm too busy and I won't be able to answer your emails in due time and manner.
Apologies in advanced.
A few guidelines when asking for help:
- Read the documentation first. Please make sure you've read the documentation before asking fo help. I know this is obvious, but a lot of people ask for questions in that are shown as examples in the documentation.
- Be polite. Again, it may be obvious, but you cannot imagine the number of emails I get saying something like
"I'm doing XYZ (which is totaly incorrect) and it doesn't work, so it' your fault because your program must have a bug!"
"You must drop everything you are doing and do my job for me right now!"
- If I don't get back to you... Do remind me if I don't get back to you within a few days. Sometimes I'm swamped with work or just out of town, a polite reminder helps.
I always take into acoount features request.
So don't hesitate to send me your ideas.
Of course, it doesn't mean I'll get all suggestions implemented as soon as I get an email.
But if a lot of people are asking for the same feature, it is a strong indicator that I should try to implement it sooner rather than later.
Please, don't be frustrated if I tell you I cannot implement your suggestion / idea. Keep in mind that some features are impossible to implement for different reasons, e.g.:
- SnpEff uses some standard (e.g. VCF) and new features cannot break those standards.
- SnpEff is used by a LOT of people in MANY diferent pipelines. New features must be compatible with current infrastructure and cannot break operational production pipelines.
- Some feature requests are just too difficult to implement