WormBase-Caltech Weekly Calls March 2016

March 3, 2016

Sentence-level indexing

• Chris playing around with flagging sentences as positive/negative for phenotype and flagging keywords as true or false positives
• It may be helpful to objectify sentences and generate unambiguous, unique IDs for sentences
• There will likely be difficulty in unambiguously pulling out sentences from re-processed papers
• Best identifier would be the sentence itself
• Sentence 'objects' could be connected to annotations in WB

Proteomic datasets

• Wen wants to systematically collect proteomic datasets
• Many are scattered across different resources
• Would also help ParaSite to index similar datasets for parasites
• MESH terms may help identify proteomic papers

MOD-human disease collaboration

• FlyBase working on a shared template for curating human disease
• SGD, ZFIN, MGI, FlyBase, and WormBase meeting at Caltech to discuss
• One priority is minimizing redundancy of effort
• There is a common portal for human diseases in progress
• Should consider Monarch initiative approach

AddGene & WB constructs

• Karen working with AddGene to pull in worm reagents (plasmids/constructs)
• AddGene does want to have better linking to WormBase

Metabolomic data

• Can we show chemical structures for metabolites?
• User (M. Witting) says metabolism people will want monoisotopic masses - these masses aren't available through ChEBI at the moment and it will be a while for them to provide them. Will find a formula and sources files for Juancarlos to implement a calculator in the OA.
• We already use, sync and collaborate with ChEBI

March 10, 2016

Adding WormBase-Parasite Paper to the Curation Pipeline (Kimberly)

• Developing new paper editor for Jane and Michael P. to approve parasite papers
• Adding species information to ALL papers; can be one or more species per paper
• Would like to then incorporate parasite papers into downstream SVM flagging pipeline and curation status form
• Species information could be used to keep curation stats clear
• We should discuss the curation and tracking of non-elegans species on the next site-wide call
• Might be best to have a separate ParaSite-specific curation status form; depends on use-case
• We may be able to just add new columns to the curation status form, Caltech vs EBI
• Michael and Jane want to curate gene function, e.g. GO; not sure what else

Dynamo-DB backend for Datomic

• REST-API now switched to dynamodb backend.
• Juancarlos could set up a form to try Datomic queries

WB FAQ page

• Need to request write access from Todd
• We should find out whether this page is used much. It needs updating considerably.

March 17, 2016

Xiaodong taking year leave

• Will try working with company in China for a year
• Xiaodong will start April 15
• Daniela taking over antibody curation, Chris & Daniela taking over gene regulation and sequence feature
• Xiaodong will discuss antibodies for disease with Ranjana

Tissue enrichment tool (Raymond)

• Should development continue with integration into WormBase? Yes, no one else is doing it
• GO enrichment tool in Intermine?

Text mining: brainstorming

• We can likely explore many more options in terms of text mining
• All possibly recognizable entities could have data filled and displayed to curators and/or end users
• What about showing PDF markup from papers in PubMed Central? There is increasing interest in participation
• We're trying sentence-level SVM for phenotype; will see how that works
• We can quote papers verbatim in annotations, but should probably use double quotes from now on
• We'll try to generate putative annotations automatically that can be confirmed or rejected by a curator or author

March 24, 2016

Outsourcing compute time

• SPELL requires 30 hours with 16GB memory
• Amazon cloud could offer 244GB memory at $3 per hour
• We already have an Amazon account

March 31, 2016

Reagent data submission form (from Nobert Perrimon)

• Data table provided as an example submission of reagent information at point of publication
• Several reagent types included
• We may want more specific information, e.g. sequence(s), relevant gene(s)
• Would we need/want other follow-up tables to fill in more specific information?
• Genetic variations should be added, including sequence
• WB curators should draw up a table of minimal information for their data type
• Chris will clean up and send around table for RNAi data submission

Proteomic papers (Wen)

• Searching title for "proteome"
• ~230 papers total
• ~187 proteomic papers might be curatable
• Many may not have expression clusters, but Wen did find new datasets