Differences between revisions 97 and 98
Revision 97 as of 2015-02-16 09:25:31
Size: 14213
Editor: ?Roland Fehrenbacher
Comment:
Revision 98 as of 2015-02-26 13:16:12
Size: 14309
Editor: AndreasTille
Comment: vsearch
Deletions are marked like this. Additions are marked like this.
Line 198: Line 198:
  * vsearch (prepared by Tim at the sprint, sorted out some issues with upstream later myself)

DebianMed 2015 Sprint

Location, Date

  • when: 30/01/15, 02/02/15
  • where: Saint Malo, France
  • Map

New Hotel location

Hotel Univers Location map on OSM

Hotel offers standard and twin (2 single beds) rooms at 50 euros per night. 11 euros for buffet breakfast.

You can book your room online. Some types of room may be cheaper via 3rd-party booking sites (eg. Booking.com)

Many other hotels around...

*NEWS*

  • The original Hotel we had planned to meet in has suddenly closed. Please cancel any online reservation done to the Bella Notte hotel.

    Previous Hotel Bella Notte

Global planning

The sprint will start on friday 30/01 from 17h00 (according to arrivals). Preparation, discussion over dinner and week-end planning will be discussed during the evening. After that, each one can work on his expected tasks/discussions... On Saturday, packaging training will be available for newcomers and help for working their own packages (or willing to maintain). The Sprint will officially last till Sunday afternoon. Several will certainly stay on Sunday evening and everyone is welcomed to join 'till monday morning for a last lunch/drink and discussion.

Venue and cost

Hotel is 20 mn walk from railway station. Saint Malo can be reached with TGV train from Paris/Rennes. You can also go to Rennes with plane or train, and take a TER to Saint Malo.

There is also a ?RyanAir airport near Dinard. From there you can take a taxi, or contact me (Olivier) to pick you up (not far from Saint Malo). Someferries from Portsmouth are available too but there are few lines at this period. See this screenshot.

Saturday and sunday breaks are sponsored.

Organisers

Olivier Sallou (see below) is the main organiser. Please feel free to contact me directly or to discuss the meeting with us via the Debian-Med mailing list.

Participants

Signing up

All attendees will need to pay in advance for the first night with a credit card to secure a hotel room and pay the balance on arrival at the hotel.

Attendees

As well as booking your room directly with the hotel, please fill your details in the table below.

No

Name

Email

confirmed

booked accom.

Organisation

Living in

Arrival at

Departure at

1

Olivier Sallou

osallou@debian.org

yes

yes

University of Rennes 1

France

30.01

02.02

2

Andreas Tille

tille@debian.org

yes

yes

Debian

Germany

30.01 @14:30

02.02 @14:50

3

Steffen Möller

moeller@debian.org

yes

yes

Debian

Germany

30.01 17:30@CDG -> late

02.02??

4

Tim Booth

tbooth@ceh.ac.uk

yes

yes

NERC / Bio-Linux

UK

30.01 @17:00

02.02 @10:00

5

Tony Travis

tony.travis@minke-informatics.co.uk

no

no

Minke Informatics

UK

6

Luca Clivio

luca.clivio@marionegri.it

yes

yes

Mario Negri Institute

Italy

30.01

01.02 @12.00

7

Sascha Steinbiss

sascha@steinbiss.name

yes

yes

Sanger Institute

UK

30.01

02.02

8

Jorge Soares

j.s.soares@gmail.com

yes

yes

Sanger Institute

UK

30.01

02.02 ??

9

Eric Deveaud

edeveaud@pasteur.fr

yes

yes

Institut Pasteur

FR

30.01

01.02

10

Bertrand Néron

bneron@pasteur.fr

yes

yes

Institut Pasteur

FR

30.01

01.02

11

Hervé Ménager

hmenager@pasteur.fr

yes

yes

Institut Pasteur

FR

30.01

01.02

12

Michael R. Crusoe

mcrusoe@msu.edu

yes

yes

Michigan State University

U.S.A

30.01

01.02

13

Goswin v. Brederlow

brederlo ([@)] q-leap.de

yes

yes

Q-Leap Networks

Germany

30.01 @24:00

01.02 @15:00

14

Navid Fehrenbacher

nfehren ([@)] q-leap.de

yes

yes

Q-Leap Networks

Germany

30.01 @24:00

01.02 @15:00

15

Roland Fehrenbacher

rf ([@)] q-leap.de

yes

yes

Q-Leap Networks

Germany

30.01 @24:00

01.02 @15:00

Agenda

(tentative/final agenda here before/after the sprint)

Friday:

  • 17h Meetup at hotel and Dinner
  • 20h Welcome
  • 20h15: Discuss on expected tasks for the sprint
    • Debian Med sprints and mentoring (Andreas Tille)

Saturday:

  • 9h30 Various presentations
    • Mobyle: H. Menager
    • ?HeavyBase: Luca

    • Qlustar: Roland
  • Hacking / Qlustar demo available
  • 19:30 Beer and pizza provided by sponsors
  • More hacking....

Sunday:

  • 9h30 Other presentations:
    • Specific uses cases of Qlustar
    • Exploring microbial communities using Bio-Linux (Tim)
  • Hacking, hacking, hacking...
  • 16:00 Compilation of activities and output reports from all hackers
  • 17:00 Thanks and close

Week-end focus:

  • osallou
    • package biomaj3, work on Mobyle2
    • investigate in Biojava 4 (request from biojava team)
  • l.clivio: current development stage and case studies involving heavybase, the peer to peer data sharing engine for clinical trials and biobanks.
  • a.tille
    • find a better solution for name space pollution (prototype done in Git)
    • check for other packages ?BioLinux might need

    • Mentoring newcomers
    • GNUmed 1.4.14 upload to unstable to reach Ubuntu sync
    • General bug fixing for Jessie
    • How works Debian Med PPA? (any volunteer for a short intro?)
    • Discuss status of sga with Michael
  • Sascha
    • discuss/finish/upload barrnap
    • discuss and work out htslib/tabixpp library issues
    • look into nextflow
    • get my key signed
  • r.fehrenbacher + Q-Leap Team
    • Discuss optimal packaging strategy when using upstream git repos
    • Setup Qlustar demo-systems for participants to play with

    • Work on new packages: mpiblast, sailfish, eman2, frealign
    • Help with bug squashing for other packages
    • get keys signed
  • Tim Booth
    • Look at BioMaj 3, and see if it can help me with rdp-classifier fungal seqs

    • Complete transition of some work-in-progress packages to Debian
    • Package QIIME 1.9 - now genuinely FOSS for the first time thanks to VSearch :-)

    • Mentor/help/learn wherever I can
    • Start a general discussion on Google Summer of Code ideas
  • Steffen
    • With Tim and Hervé report on the ELIXIR catalog hackathon in Copenhagen, still hoping for Kristoffer/Jon to make it to the meeting
    • Brainstorm about milestones for Debian Med and future Sprints
      • presentation as workflow-infrastructure rather than as package collection
      • tie up with Bio* trainers?
      • shall there be Sprints in other parts of the world? Or Mini-DebianMed-Confs?

  • a.tille + ??? (volunteers wanted!!!)
    • Working on Jessie release notes
  • Michael R. Crusoe / @biocrusoe
    • packages for our software (khmer & screed)

    • packages for other apps we use: prodigal, trinity mRNASeq assembler, RSEM (RNA-Seq by Expectation-Maximization), EBSeq
  • Bertrand Néron
  • Hervé Ménager
    • Present Mobyle
    • work on Mobyle2
    • work on CWL -> example from debian?

    • present/work on Elixir registry -> example from debian?

  • Éric Deveaud
    • Learn to package for debian
    • package softwares from my lab (toppred and prose)

Reports

Personal Reports

Olivier Sallou

  • Vienna-RNA: fix licensing issues to upload in experimental
    • Waiting ftp-master acknowledge on specific license point
  • Mentoring on macsyfinder with Bertrand (ready for inject on svn once ssh key is accepted and upload sponsor)
  • BioJava 4 : package injected in svn, and uploaded

  • Logol: prepare patch to avoid removal due to gridengine dependency.
  • biomaj3: prepared package for up-to-come release, not yet injected in svn
  • Discuss on gridengine package with Clustar people

Andreas Tille

  • Package uploads
    • rna-star (uploaded to new)
    • plink (manpage in debian-med scope)
    • python-screed sponsored on behalf of Michael
    • toppred (in the mentoring process of Eric and Herve)
    • GNUmed server v19.14 + GNUmed client 1.4.14
    • sphinxcontrib-autoprogram sponsored on behalf of Michael and Tim
    • vsearch (prepared by Tim at the sprint, sorted out some issues with upstream later myself)
  • Name space pollution issue:
    • Blends Git has a solution for bash users. A script /etc/profile.d/debian-med.sh will be installed which evaluates users ${HOME}/.blends file. If it contains the string "debian-med" the user gets an extra dir prepended to PATH that contains the originally named binaries.

    • Enable also manpages in Debian Med scope
  • Small intro talk about Teamwork in Debian Med

  • Work with Michael and Tim on python-screed and khmer (and help getting them a cowbuilder sid environment under Ubuntu)
  • Mentoring of Eric and Herve at the example of toppred (up to final upload)
  • Fix some Authors strings in several upstream metadata
  • Checkout pre-dependencies for freebayes and sga (libvcflib, libtabixpp, python-ruffus) which were topic of last sprint but never finished
  • Work on some further preconditions
    • ngs-sdk (needed by new version of sra-sdk) - build system is a mess
    • python-ruffus (needed by sga) - some parts of the test suite are failing, need to contact upstream
    • try to update sra-sdk build (build system same mess like ngs-sdk)
    • try squashing bugs in vsearch
  • Draft for Jessie release notes
  • Render teammetrics graphs for January 2015
  • Some SoB sponsering for Debian GIS and Debian Science
  • Enhancing khmer packaging prepared by Michael

M. Crusoe

  • Work on screed, khmer, prokka, trinity
    • screed: in NEW queue
    • khmer: waiting on sphinxcontrib-autoprogram; then submit
    • prokka: research licenses on databases, check package status of dependencies
    • trinity: >50% done, Licences need auditing, Java muck

  • GPG key created, signed; signed others
  • common-workflow-language tool definitions review with Herve

S. Steinbiss

  • barrnap, barrnap-silva-hmms, genometools
    • barrnap: license issues with included HMMs prevent inclusion of whole code into main 23S and 28S rRNA profile HMMs are based on SILVA alignments (only free for academic use)
    • SILVA contact person (Gloeckner) confirms this is also the case for derived data scripted repacking to remove the questionable HMMs from set of files patched barrnap executable to consider both free (in this package) and non-free (in barrnap-silva-hmms) HMMs for the nhmmer runs barrnap-data-nonfree goes to non-free/science contains the non-free HMMs in a separate directory
    • aragorn: added EDAM annotation
    • genometools: add EDAM annotation (it’s a lot, may take a while)
    • REAPR: decide about tabixpp: package it with ‘old’ tabix (pre-htslib) embedded? there’s no other way, htslib interface is completely different

no license for tabixpp -> contact upstream (Erik Garrison)

Tim Booth

  • Merged my khmer packaging work into MCs stuff resulting in packages suitable for Debian upload
  • Attempted to assist MC on Trinity packaging
  • Started on packaging QIIME 1.9
  • Reviewed licensing on VSearch test files as requested by FTPmaster
  • Gave presentation as planned
  • Had a crack at running QIIME on Qlustar, and reminded myself how much of a PITA QIIME is, but made plans for progress.
  • Looked at/discussed amendments to ELIXIR registry integration with SM and HM.
  • Got a demo of Biomaj3 from Olivier

Herve Menager

  • Work on EDAM annotations
  • Package training

B. Neron

  • Package training
  • Work on macsyfinder, almost ready for upload

Eric

  • Package training
  • Start packaging of a program

Q-Leap team

  • Qlustar demo and presentation
  • Started packaging of libshark (almost ready to upload)
  • Started packaging of g2log (almost ready to upload)
  • Started packaging of sailfish (almost ready to upload)
  • Started packaging of mpiblast
  • Discussed the requirements to get QIIME running on Qlustar and setup a test environment for it.
  • Key signing

Acknowledgements

the sprint has been possible thanks to:


CategorySprint