| | 1 | = October 2012 Evaluation GoNL = |
| | 2 | People present: Morris Swertz, David van Enckevort, Paul de Bakker, Lennart Karssen, Kai Ye, Tom Visser |
| | 3 | |
| | 4 | E-mail contributions: Hailiang Mei, Jan Bot |
| | 5 | |
| | 6 | == Introduction == |
| | 7 | |
| | 8 | |
| | 9 | == Action items == |
| | 10 | |
| | 11 | |
| | 12 | |
| | 13 | Technical: File management & replication |
| | 14 | * General backup strategy and restore? |
| | 15 | * What is where (ToC of files)? |
| | 16 | * Is the file in hand the same as in ToC (checksum)? |
| | 17 | * What version is this file (e.g. multiple align runs) |
| | 18 | * Does the researcher have the file available on site? |
| | 19 | * Data freeze: can we mark data sets. |
| | 20 | * Data librarian: who is responsible for keeping the lists |
| | 21 | (dcache instance writes) |
| | 22 | |
| | 23 | Action items: |
| | 24 | => create a series of user stories describing |
| | 25 | => Version individual files, not the whole set because to big (+index, etc) |
| | 26 | => Have overview of who wants what |
| | 27 | => Small files we can release as a whole, e.g. SNP releases |
| | 28 | |
| | 29 | Technical: Distribution of the analysis |
| | 30 | * Where do you compute what? |
| | 31 | * Can we really distribute analyses over multiple sites |
| | 32 | * Currently we depend on LISA and UMCG clusters. |
| | 33 | => Make pipelines distributed: deploy pipelines on multiple clusters |
| | 34 | => Make dependent executable available on other clusters |
| | 35 | => Make data available on other clusters |
| | 36 | What pipelines do we want to distribute and why, and what are the barriers??? |
| | 37 | |
| | 38 | Technical: QC and tracing of errors |
| | 39 | * Robustness of the analysis |
| | 40 | * How do we make certain that data analyses are used |
| | 41 | => Action item: clear QC steps but pragmatic. E.g. compare unique aligned reads. |
| | 42 | => Action item: verification of pipelines accross sites using overlap samples. |
| | 43 | |
| | 44 | Coordination: Communication problems |
| | 45 | |
| | 46 | |
| | 47 | Organization: which resources are actually available |
| | 48 | |
| | 49 | Science / Roadmap: |
| | 50 | * Paper plan |
| | 51 | * Get from the steering committee general directions, very broad, what |
| | 52 | can / should do next with the data (GoNL flag, or just using) |
| | 53 | * Group responsible of rolling roadmap for one year (get from the |
| | 54 | steering committee) |
| | 55 | * Have more bioinformaticians in the steering committee and recognition of that |
| | 56 | * At every SC meeting have one of the subproject report results to SC |
| | 57 | * Overview of external GoNL projects |
| | 58 | * Very good that we have a SC member (Cisca) on the call all the time. |
| | 59 | * The technical people should get appreciation for their scientific |
| | 60 | contribution! |
| | 61 | * Need experienced person for each working group (SV is okay, |
| | 62 | imputation and pheno are a bit light because Yurii left) |
| | 63 | * Foreign contributors is nice, but it seems like they take away nice |
| | 64 | projects away. Need better communication. |
| | 65 | |
| | 66 | Organization: Roadmap and planning |
| | 67 | * Who is responsible for what? |
| | 68 | * Decentralized management (we can not boss other locations) |
| | 69 | |
| | 70 | Actions: |
| | 71 | * Ask the SC people resources available (do the GoNL members get the |
| | 72 | time they need?) |
| | 73 | * SV team has too little man power to do the work (largely volunteers, |
| | 74 | hard to stimulare people) |
| | 75 | |
| | 76 | Keep: |
| | 77 | * Weekly skypes |
| | 78 | * Mailing list |
| | 79 | * Open communication and low-threshold to find each other |
| | 80 | * Sharing of best practices nationally and internationally |
| | 81 | * Forming the group |
| | 82 | * Access to international collaboration |
| | 83 | * Sharing knowledge and code via wiki+svn |
| | 84 | * Self-organization in working groups along sensible lines |
| | 85 | * Using pragmatic solution and get started |
