2013 04 19 Teleconference Minutes

The minutes from the April 19, 2013 COR call are below. Please log-in and update the document if you have any of the information in red text.

Databases and Ontologies CoR Teleconference

Minutes

19^th April 2013

Participants:

Nathan Baker (Pacific Northwest National Lab), Hubert Rauscher (JRC), Stacey Standridge (NNCO), Karen Griffin (University College Dublin), Neill Coen (University College Dublin), Christoffer ?berg (University College Dublin), Iseult Lynch (University of Birmingham), Fred Klaessig (Pennsylvania Bio Nano Systems, LLC), Tim Booth (CEH), Victor Maojo (Universidad Politécnica de Madrid), Jessica Adamick (University of Massachusetts Amherst), Christine Hendren (Duke University), Wouter Fransman (TNO), Rafi Korenstein (Tel Aviv University), Joe Fisher (U.S. Army), Krishna Rajan (Iowa State University), Teresa Fernandes (Heriot Watt University), Diana de la Iglesia (UPM), Lauren Neily (Nanosafe Blackburg in Virginia), Marty Fritts (SAIC-Frederick, Inc. / NCI at Frederick), Miguel Muñoz (UPM), Agusta XXX? (XXX), Vijay Gupta (RTI), others?

 

Introduction from Nathan and Hubert

Presentation from Christine Hendren from The Risk Assessment Core:

Aim: to communicate what we need in the Risk Assessment COR:  what do we need from other CoRs and what do we need to provide to others.

 

Slide 2: Purpose – information to action continuum – ultimate goal of all CoRs is to integrate data and have that inform decisions regarding risks – illustrates the links and roles within the CoRs

Understand what value we as CORs are adding, and what are the low handing fruits that we can address with limited funding.

Slide 3: Ontology and databases are foundation for the whole knowledge pyramid:

Information: open literature, data sets etc., so ontology and database CoR can organize and subsume this data, and facilitate the other CoRs to utilize the data.

Other CoRs draw on the increasingly organized data to make predictions regarding exposure, transformations, etc. for their models and forecasts.  Finally, the risk assessment CoR takes all this data and interprets and sysnthesizes it into risk forecasts and practical risk management tools that are utilizable.

Idealised risk continuum – doesn’t necessarily translate into outputs.  Trade-offs

Slide 4: shows the same pyramid to frame what risk assessment COR needs.  Risks = exposure x hazard.

How the CoRs provide the different pieces of this equation, and risk COR needs to frame these to arrive at a risk forecast for NMs.  Intelligent consumption of knowledge.  Articulating these needs and trade-offs up front – continuous feedback required – organizing the data at the same time as categorizing and learning.

Articulating the idealized hand-offs between these communities will be a useful resource and can support the closer integration of approaches and strategies in EU and US.  Hopefully will support the CoRs overall.

Mark Hoover and Christine put this into graphic form, but all CoR participated.

 

 Q&A for Christine:

Nathan – use future calls of the group to further articulate this approach and continuum.  Is there a wish-list or top needs that might drive the interactions. E.g. where might open literature or databases have information that you need to exploit?

Christine: Currently making lists of current resources, and what is known versus what is needed to know to make risk forecasts.

Exposure, ecotox and human health –important to categorize the systems in which the data has been collected – difficulty drawing a risk assessment conclusion – e.g. deionized water versus specific freshwater –

Nathan: will you share the datalist with the group?

Risk COR meeting is scheduled for next Tuesday (23^rd April 2013), and will see how much progress has been made. 9 linear objectives that build upon each other.  Complete open sharing is Christine’s preference.  Would make sense if people want to call into the Risk CoR call.

Stacey: All COR information is on the website. Also have a preliminary list of relevant papers for the risk CoR – all additions to that are welcome.

Marti Fritts: open literature and data – did you also discuss sharing models as part of the infrastructure?

Christine: this is included conceptually, but will specifically incorporate in slides – good feedback.

Next steps:

Point the Risk CoR to the Databases CoR’s lists of data and examples of where additional data / protocols etc. would be important will help us to firm up what we are doing / planning.

 

Presentation from Victor Maojo from InfoMed:

1st real paper in area of medical 1959 Science paper – medical diagnostics – took at least another 20 years to get a consolidated discipline. E.g. human genome project.  Needs a BIG Goal / big project to drive this consolidation.

Many topics similar across medical informatics, bioinformatics, nanoinformatics – many similarities but many things different also – several things need to be adapted –semantics etc.

Bioengineering, bioinformation, nanoninformatics – overall vision needed.

What is the opportunity for nano-informatics?

Different perspectives – don’t only build computer applications- needs also to be scientific – people who try to gather and reason in between different inter-disciplines.  Behave as brokers between the areas – not only engineering.

Project in Victor’s lab: built from a pure nanoinformatics approach, and the challenges.  Text mining: automatically extract information from different resources, e.g. ISI, extract information from literature etc.  Structured or non-structured information.

ð  Nano-resourcosome – big effort over 6 years – medical informatics and now nano-informatics/ BMC bioinformatics and BMC

First overview of inventory of nanoinformation – extracted information from XXX papers?

Ontologies efforts:  what are the challenges? would like to add an expanded view – how to expand and how to link classical text-based conceptual information linked to graphical visual information in order to create a new type of conceptualization.

E.g. NPO – shapes of nanoparticles: all classical ontologies only deal with shape with textual information, but this loses the graphical information that is critical.

Preliminary prototype linking NPO to foundational model of anatomy – very separated in terms of biomedical levels, were able to build the nanoparticle toxicity searcher – new England journal of med 2010 –

These tools could be used to search for more information to find literature related to NP and brain toxicity.

A corpus for nanotoxicology – if you have students that would like to do work for 3-5 days and then publish a short time – read it later- point students to this? E.g. email Victor!

1st prototype of visual taxonomy of shapes.  Student Miguel (also on the conference call) – could be type of this overall target to build a nomenclature of nanoparticles, but cannot be isolated from morphological-spatial information regarding the particles.  Once paper is submitted will share all but for now part of several MSc theses so cannot be public just yet.  Capture and automatically classify particle and structure shapes with textual and visual information.

New tools for shape extraction / description: Very important to create a geometrical shape. spatial distribution / descriptors. Outside of text-based (conceptual) ontologies.

Automatically extract data – link classical approaches

Presented informatics ideas to European Commission: glad to get some proposal for nanoinformatics for ICT and DG-connect. Would like one concrete proposal for this topic.  Submit a short proposal in 1-page regarding what community could carry out this work – ideas for the next WP for H2020.  Glad to analyse this proposal topic:

 

Q&A for Victor:

Hubert: Build a standard corpus using students from different groups:  Have published an early prototype  – need the collaboration of nanotechnologiests as it was so-far done only by informaticions.  Collaborate to expand and make into a real toxicity searcher.

Marti Fritts: Stimulates a lot of thought.  Work on shape was very interesting – how do you exchange models for materials? We are well past the stage where we should have been expanding on the BDP – exchanging materials models on the basis of shape? As well as the molecular modeling that is already in the laboratory:  Focus the attention on how to use those models in risk modeling.  What does morphology have to do with shape and function?

Victor: PDB can hold the general information.

Nathan? More on the medical informatics side: HL7 – exchange of information in medical information. Has is expanded into nanomedicine / nanosafety?

Victor: not expert in this, but don’t think anyone has tried to expand for nanomedicine yet. Probably this depends on the applications: if there are real clinical applications they will be addressed by standards.

 

Other CoRs Business:

Ongoing efforts to collect existing sources of information:

Nathan & Hubert:  please comment and revise the current draft of the living document  – important that we could talk about a revised version of this at the next meeting.

Thoughts on the new meeting format?

Are people interested in continuing the idea of speakers both from other CoRs or from research perspective?  Agreed to keep this format.

 

Actions: Contact either Nathan or Hubert with suggestions.  Slides to be posted on the CoR website

 

Thanks to all and will schedule the next call shortly.