wiki:PombasePlanningMeetingMinutes20130429

Pombase Planning Meeting Minutes 29 April 2013

Cambridge - Sanger Building Seminar Room 2-4 pm
Present: Midori, Jürg, Antonia, Mark, Dan, Steve
Remote: Kim
Chair: Midori
Minutes: Antonia

New action items from April

  • AI: Document version naming for gene model and annotation updates (Mark)
  • AI: User documentation on version naming for website, should be 'generatable' from Marks documentation (Midori)
  • AI: Automatic updates for the FTP site; script-as-you-go-along with the updates (Mark)
  • AI: Block scary error messages on www from user’s view (Mark).
  • AI: Take a look at the biogrid interaction data and see if some human-friendly text-linkages that imply directionality can be made between interactor A and B (Mark)
  • AI: Curators and Kim to record curation stats. Should separate staff & community.
  • AI: Curators to ask community for pictures to rotate on the front page.
  • AI: Curators to flesh out what to do at the Tuesday lunch-time community curation demo and Thursday lunchtime PomBase? demo at Pombe13

Outstanding Action Items

  • Organise a large general "PomBase" banner – sort out with people doenstairs in Biochem. Go for a long skinny one with PomBase? logo and same color and yeast images as on PomBase? banner (Midori)
  • Curators to draw up a script for the curation video. What steps do we want to show on the video? Discuss at next curator meeting.
  • Remove ':pep' from identifiers in cDNA FASTA file (Mark or Dan)
  • Mark will do general PomBase poster for Pombe 2013
  • Midori to look into beer mugs/keyrings and T-shirts (in progress, see minutes)
  • Mark to work on automated release pipeline with regression testing. Plan is to have a pipeline ready for the next meeting (Mark - In Progress, see minutes).
  • Load and display RNA/protein expression data from Sam's paper in V. 35 (Mark - In Progress, see minutes)
  • Mark to sort out GO slim count anomalies (see PB-1208)

Postponed Action Items

  • Demo of curation tool (postponed)
  • Val/curators will start to send out a large number of community curation sessions (postponed until new session management and documentation is in place. Will continue to send out small numbers of papers) - getting closer
  • Mark to liaise with Giulietta to help with making a Pombe community curation video.
  • Dan and Mark to see what files downloadable from PomBase have previously been generated by ensembl. These should be easy to create automatic updates for (In progress?)

New issues

General

Update cycle

Running updates are now ok. GO, FYPO, pretty much automated. Lots of little scripts do individual bits. To go from first script download chado database, takes within a day. Then it updates each chromosome on the cluster; all 3 can be done at same time. Doesn’t update the MART. The MART stuff needs to be updated manually at the moment. Problem found was that that only updates annotations, not gene models. Now the pipeline goes to write the whole thong as if starting with no database. This will be done each release. Between releases there will just be annotation updates. Mark now has scripts to show if coordinate changes have been done (Mark). Having an independent gene build version would be useful for some users (Dan). From curators POW we will be ok if gene model changes happens at the same time as ensembl genome releases. We want to know which ones include new ensemble releases (Midori). I didn’t know there were changes in the gene models (Mark). There is room in the ensemble schema for tracking gene build version (Dan). Call it something like 33.1.1. One thing you can do is that the major version will be the gene build, incremental versions for annotation (Dan). Assembly all taken care of by ISDC (?) (Dan). We need to document this and make it easy to find (Midori). Ensembl genome version should be kept out of this numbering (Dan). “1” will be the feature set inherited from GeneDB. 2 will be the one pushed out now. 2.1 -> no UTR or exon changes, just annotation changes. AI: Document version calling (Mark) AI: User documentation for website, should be copy-pasteable from Marks documentation (Midori)

  • Now showing v33 (7th March; 535 fully curated publications)
    • Chado load done for v34; live yet?

Mark has updated on his local box. Should be on stage soon. Also Sam’s data. Will be showing on gene pages.

  • Data ready for Kim to do v35 chado load

This is fine from curator’s end. Mark will go on holiday so it is not imminent this is done for bank holiday?

  • Need ftp site update ASAP after v34 live (even if some has to be done manually)

Must be done after 34. For the files that are in the same directory, they can be copied over easily, but I haven’t scripted it yet (mark). Script as you go along with updates (dan). A lot of the outstanding stuff on jira has to do with various things where a version exists but needs updating. Some things are not offered yet by FTP but we want them (midori).

  • Add Jira fix-versions for next couple chado releases

Chado release numbering should stay the same. Useful for us curators to know how fast the chado releases move through the pipelines.

Usage stats

http://tinyurl.com/7qs2zap

Users come from, in order of volume: US, Japan, UK (Mark). Japan has the largest number of labs (Jurg).

Website

  • Recent troubles with references & CiteXplore/EuropePMC
    • How confident can we be that missing references or unresolved appearance (PMID:000 vs Author et al. yyyy) problems won't recur?
  • SOAP (and other) errors appearing on gene and static pages, likewise

Spikes came from our server and triggered them to slow down. So they limited us. We now have changed how our access works. We now do an access and a microsecond wait until the next query (instead of query ->no response ->new query which becomes a vicious loop). We have separated us from the EBI proxy. Mark and ES sorted that. So now the scary messages will be gone from the website? why were they on static pages? (midori) The error was genereated while querying, the error message gets stored for subsequent pages (mark). Is it optimal for the errors to show? (Midori) They can be blocked. (Mark). AI: Block the scary error messages from users.

  • Sam's data - in Chado v34; will it/did it go live with rest of v34?

See discussion above, this will happen.

  • GO slim counts (PB-1208)
  • Can we decouple static Drupal page updates (dev --> www) from data updates, so the former don't get delayed if there are problems with the latter? (PB-1270 done, for which thanks; but what about the general case in future?)
  • ETA/effort estimate for reciprocal interaction annotations (PB-873)?

This is by inferring directionalities. There should be a way of inferring directionality from the database. This is in the feature relationship table. (Kim). Problem: biogrid just states interactor A and B, not bait prey etc. Might be complicated. Could we simply provide some helpful bits of text such as A suppresses B etc this might suffice for now (Midori) AI: Take a look at the interaction data and see if some text-linkages can be made between interactor A and B (Mark)

Other general issues

  • Anything to report on incuding PomBase in EBI site-wide search?

This is done (Mark)

Curation

  • Literature (triage) status
ItemMarch meetingApril meeting
All publications97559761
Un-triaged publications04
Curatable publications47354740
Publications with approved sessions580600
Publications with active sessions245247
Publications with session needing approval144
  • Canto annotations (from 2013-04-26)
nameMarch countApril count
PSI-MOD7777
molecular_function167173
cellular_component211245
biological_process838893
PomBase annotation extension terms14481605
fission_yeast_phenotype35193820
Total62606813

I think we record the date when annotations are made so we can easily get some stats. Perhaps a monthly one would be a good start. Just the curation tool? (Kim) Lets start with that. It gives us a handle on how we are doing. Staff vs community distinction (midori) AI: Curators and Kim to record stats. Staff vs community.

Phenotype ontology

  • 2016 terms
  • distinguish cell viability and population viability

Will it be possible to have allele filters in the query builder (Antonia). This will be possible but not for pombe meeting. I will need to look at it (Mark).

Next priorities

Some easy some difficult (Mark). Mark has some ideas with both Pro’s and con’s. But mapping should be made at the chado end so it should be very easy. Maybe.

We need something nailed down with regards to what it will look like (Dan). What will be in second column? (Mark), Keep existing header but basically how SGD looks. News on right, pictures on left. About PomBase? on left. Pombe meeting on left. Community curated stuff on left – link to publication pages (but this is probably future? Could it be done in time? Might not fit in priority list)? We could just put the citation for now with a citation links. We can update this with PomBase? publication page link later when there is time. Mark will do this on stage so we can look at it. AI: Curators to ask community for pictures to show on the front page.

Should be easy because analogous to other stuff (Midori). This is not loaded into the ontology database yet but shouldn't be a problem (Mark)

I don't think I can finish that for the meeting (Mark)

Publications

FYPO paper coming along in the foreseeable future.

NAR update paper needs to be in for September. Request beginning of June/July? (Mark).

Manuscript on Canto in progress, no first draft yet. Will go to bioinformatics application note or something like that.

Meetings

Future

  • Pombe 2013
    • 13 May curator meeting + curator/programmer conf call devoted to planning our activities, esp. lunchtime demos, tutorials, etc.

1.5 hrs lunchtime so 45 minutes could be devoted (Jurg). One on Tuesday for community curation and one on Thursday for the website but the programme can be changed up until early June when the programme will be printed. This will be discussed on May 13. Should Mark be present in person? Probably skype.

  • Midori to look into beer mugs and T-shirts

Too expensive to get for everyone. Smaller number for people who curate papers? Smallest min order Midori seen is 48, no quote yet. Not handy for people to bring back. Keyrings? Lightweight, cheaper. T-shirts; Jurg thinks we should wear onesies. Curators will discuss this more at next meeting.

  • ideally would like Canto tutorial video ready

Next planning meeting

Wednedsay 29 May, 2-4 pm, at EBI

Last modified 7 years ago Last modified on Apr 29, 2013, 5:21:42 PM