==== January 2016 Notes ==== |~~TABLE_CELL_WRAP_START~~ **Participants** Nils, Leonid, Job, Chuck Murry, Lisa Postow, Larry Gerace, Stephanie Alexander, Ewan, Olivier, Thimothee Lionett, Lisa Chadwick (notes) - **Scope and mission: Publications, data, software** - Goal is to create a policy on publication, data sharing and data release, and software/algorithm sharing. - Policy to cover both internal and external sharing. - Updated the mission language directly on the wiki - **Meeting frequency** - Monthly calls - Nils will send out another doodle to find a time, but the Friday noon ET might work. We will start with Feb 5. (first Friday of month) - If there are more frequent discussions needed, we can schedule that. - **Communication possibilities** - Mailing list - Wiki - Shared documents (google doc) - We will use google docs (linked through the wiki) to do most work. Announcements through the 4DNPUBPOL@list.nih.gov - **Experiences with other consortia** - Ewan: assumption of high transparency within the consortium is important. - Leonid: a system that encourages exchange of ideas and sharing, but also provides protection - that you get credit where appropriate? Not just data, software, but also intellectual property.\__ - Olivier: Documentation of sharing (i.e. so and so gave a presentation on this on a certain date) - Job: a shared document (google doc) where people disclose papers in planning. You would list it as soon as you were even thinking about it. All relevant publications: not just what is directly funded by this project.\__ - Ewan: it requires constant reminders to get people to do this. - We could create a shared document with a list of abstracts/information about papers in progress. People can identify areas of potential overlap and work together to come up with a solution. - You could publish a paper on your own paper with your own schedule, you could coordinate with another group. Consortium-wide analyses required more coordination. - Social dynamics is important! - We can create a list of abstracts, titles, etc. Remind them to update it on SC calls, etc. - Title, authors, abbreviated abstract, datasets used, datasets and software that will be produced. Data from more than group - they should all be authors on the paper. - An automatic notification system would be good for this. Lisa will talk to the OH about this. - Leonid: Papers submitted: at what point should they be shared with the consortium? When it is accepted? Bioarxiv: preprint server. Its is up to you when you update the paper. Maybe share paper when submitted, and when accepted. We put our papers on bioarxiv when we submit them anyway, but we could just do this within the consortium. It also helps provide an additional level of peer review/discussion that might be useful. Ewan: I support the sharing of submitted and accepted publicataions but I am not sure how much the commenting system would be used. - Writing down the rules is important, but in person interaction is also important. A high level of trust needs to be established. - The ENCODE policy and conflict resolution mechanism is good and seems reasonable. We will model off of this. - Shankar also mentioned data release policy. Metabolomics policy: data should be released as soon as the paper is accepted. We don't want them to wait 5 publications before we release the paper. It is bad scientific practice to not release data when the paper is out. - ENCODE had more of a community-centric view on data release. We do not necessarily need to do that. - Remember that we are talking about not only sequencing data, but also imaging data. The communities do not necessarily have the same mindset/rules. - The omics/imaging standards groups should be discussing what data exactly should be shared with the DCIC, that will be a start for determining what data should be released publically. - Define a minimal standard. - Stephanie: Imaging data - submit at least a sample dataset that has been analyzed, as well as the software that was used to analyze it. That is currently required by many journals. \__ - We will discuss data release (when should things be released and under what conditions) in a future call. - If anyone is aware of another consortium's software sharing and release policy, please share it. Ewan: NHGRI large scale sequencing programs might have had some software sharing. Shankar: Alliance for cell signaling, lipid maps projects - he will try to find policies and share. Job mentioned that ENCODE was trying to do this. One question was: who is responsible for updating the software over time? ~~TABLE_CELL_WRAP_STOP~~|