2013 CTSI Annual Pilot Awards to Improve the Conduct of Research

To facilitate the development, conduct or analysis of clinical & translational research

Information Interface for Patients, Clinicians and Researchers

Proposal Status: 

Rationale:Taking findings from basic research to practical applications that enhance human health and well-being is the signification of translational research. Improvement in human health and well-being, however, does not necessarily involve the cure of a disease. Sometimes mere information transfer might help patients substantially in dealing with their medical condition. This is particularly true for patients with family history of yet unknown or contradictory diagnoses – in other words for patients with rare hereditary diseases. This point was emphasized very recently by Erika C. Hayden in Nature [1]. Providing information is the foundation of molecular genetic diagnostics and genetic counseling, which should readily be available for any inherited disease, even if no cure will be available in the foreseeable future. The provided information can answer basic questions of the patient, such as What do I have? Why my family? Is there a risk for my children? Is there anything I can do about it? And this will assist the patient in understanding their medical condition.

COL4A1 (OMIM: *120130) and COL4A2 (OMIM: *12090) encode for extra cellular matrix proteins that constitute basement membranes. Mutations in COL4A1 and COL4A2 cause multi–system disorders including porencephaly, cerebral small vessel disease with hemorrhage, Axenfeld-Rieger anomaly with glaucoma, and variable muscular dystrophy including muscle-eye-brain disease. To date, more than 50 different pathogenic sequence variances have been described in highly penetrant multi-system disorders. For ICH – the only disease so far that has been systematically screened for COL4A1 and COL4A2 sequence variants – we estimate that mutations in COL4A1 and COL4A2 may cause up to 10% of all cases of spontaneous ICH. The Gould lab identified the first COL4A1 mutations in mice and humans and recently published a comprehensive review on the topic. For these and other reasons, the Gould lab receives inquiries from researchers, clinicians and patients throughout the world interested in collaborations, clinical advice or participation in research. We recognize that many other labs at UCSF and around the world face these same types of requests. We propose that developing a centralized information management system (IMS) to promote and streamline interactions and communication between these three stakeholders will be an efficient and effective paradigm to enhance translational research.


Based up on the platform developed by the Leiden Open Variation Database (LOVD) we will generate a centralized information management system (IMS) that streamlines interchange of information between patients, clinicians, and researchers. The IMS will be tripartite in order to provide the relevant information to the appropriate target audience. We will develop the platform using COL4A1 and COL4A2 as a scalable model for any number of other groups.

  1. Patients will find a comprehensive overview on the different disorders associated with COL4A1 and COL4A2, as well as a platform to connect to medical doctors, researchers, and genetic counselors especially trained on these medical conditions.
  2. Medical doctors and genetic counselors will find a locus-specific database that collects all sequence variations in COL4A1 and COL4A2 with links to the according reference of their initial description, the latest literature on COL4A1 and COL4A2, as well as information on ongoing patient studies.
  3. Researchers will maintain the IMS and thereby find a useful research dataset for molecular diagnosis, large-scale mutation statistics, and the determination of genotype-phenotype correlations. Research will also benefit from a collaborations platform, which promotes discussion in the field. They will also identify patients willing to donate biological material for further investigations on multi-system disorders.


Criteria and Metrics for Success:

Immediate goals:

  1. Set-up a locus-specific database for COL4A1 and COL4A2 based on the format used for the Leiden Open Variation Database (June 1, 2013).
  2. Set-up a literature database on COL4A1 and COL4A2 based on the Mendeley open source reference manager (June 1, 2013).
  3. Write a comprehensive overview on the different disorders associated withCOL4A1 and COL4A2 and update all according NCBI datasets (July 1, 2013).
  4. Collect information on MDs and genetic counselors that are especially trained on medical conditions associated with COL4A1 and COL4A2 based on information provided by experts and databases, such as GeneTests(July 1, 2013).
  5. Create an interactive web-interface that links all the different resources by July 1, 2013.
  6. Go live by August 1, 2013.


Mid-term goals:

  1. This IMS will increase clinicians’ awareness of COL4A1 and COL4A2 as one possible cause for multi-system disorders and they will find necessary resources and information to help make decisions for their patients.
  2. Interested patients with multi-system disorders can arrange to donate biological samples, such as DNA or skin biopsies, to researchers listed on the IMS in order to promote research on multi-system disorders.
  3. The IMS can be a model for other inherited diseases and may lay the seed for an extended IMS that comprises also other genes and diseases.


Long-term goals:

  1. Once the platform is established we will use it as the model and continue to build this resource for other inherited diseases of the extracellular matrix.
  2. Establish this platform as a template for other UCSF researchers and clinicians to interface with each other and with patients to promote information transfer and patient recruitment.
  3. In summary, this proposal will improve the conduct of research on a devastating disease here at UCSF and worldwide.


Total Budget: $22,000

Cost estimations:

Summer student: $4000/mo for 3 months = $12,000

Web Developer: $50/hr for 1 month = $8,000

Postdoctoral Fellow to oversee project 15% effort for 3 months = $2,000



1.            Hayden EC: Data barriers limit genetic diagnosis. Nature 2013, 494(7436):156-157.



I am a librarian at UCSF and wondered if we could provide any help to you in developing a search methodology to insure that the latest material about your topic is discovered and collected for use in your IMS.

Let me know if we can be of any assistance.


Whit, we greatly appreciate any input. At present, several members of the lab use appropriate search terms with searches such as PubCrawler or Journal Lab to identify newly available information. However, we are always open to other search methodologies that may be useful. Aims of our proposal also include the implementation of a shared publications database using MENDELEY and the optimization of the “online presence” of the relevant diseases.

This is a very interesting idea that addresses a vitally important need, especially given how close we are to interrogating patients' genomes as a routine part of clinical care.  Could you explain a bit more about how you see patients using this platform?  Will MDs and genetic counselors volunteer their time to answer questions raised by patients and family members?  Will patients have the opportunity to submit samples for genomic analysis?  And if they already have their exome or genome in hand, will they be able to get help interpreting it?

 Thank you Dan,

We essentially aim to curate, streamline, and optimize information on the relevant diseases that is either missing, or already present from disparate sources (including OMIM, Wikipedia, GeneReviews, GeneCards, and GeneTests) and provide a central gateway to access this information to patients, physicians and researchers. By this, search engines will score the relevant web pages higher and patients can access the relevant information more easily. Patients can link to GeneTests where they may locate centers that already perform genetic testing and provide genetic counseling (and these would be updated as others come on board). Patients interested in submitting samples for research purposes would have the opportunity to link out to our lab page (or others) that accept samples for research purposes. Presently it is impractical to offer interpretation of individuals’ genome or exome information in such an interface as this but one may imagine this being a place to find such information in the future.

Interesting proposal.  This is one though that it will be key to understand what the landscape looks like.  For example, some kind of service such as the one you propose is part of 23andme's offerings - it needs to be, since they are in the business of providing people with genomic information.  They keep their customers coming back by providing all kinds of nuanced information, new relevant publications, expert clinical and scientific opinion, etc.  They are of course a private company and I couldn't find any indication that they provide this generally to people.  I agree that a more public service would be valuable.


But the other piece that needs to be addressed is just how this effort scales. At this time it reads very effort-heavy and is hard to imagine it scaling. It is a pilot, so in this first phase may be fine, but it would be best if you could share how you're thinking a scaled version might be implemented.  So thinking through answers to Dan's questions will help.  In addition, will be useful to review (just maybe list) some of the existing services providing at least components of the initiative you propose.  Then you can describe how you'd stitch together already existing resources and add the components that are missing.  For example, there are resources such as this one  http://www.ncbi.nlm.nih.gov/books/NBK7046/  - could you re-use content? perhaps a partnership strategy (don't have to implement now, but would plan for it for the scaled version). Other groups to think about include 'patients like me' (I don't know if they do anything in the disease risk space), or even large foundations.   

Thank you for the comment. We would hope that this becomes a kind of portal for all to use and not only those who pay for a commercial service. You raised very good points about scalability. We propose that our effort will work as an example. As part of the scalability we will provide a summary to CTSI and other members of the UCSF campus about our experience and a step-by-step protocol on how to improve the “digital presence” of disease–based research using similar methods. Information would include a checklist of suggested hosted databases to update and link. If there is interest, we can share also our knowledge on the LOVD and as a long-term goal if many other labs want to put their data also into this repository for genetic variants, one could spearhead a common LOVD database for UCSF.

This is an interesting idea.  I agree that it may be difficult to scale.  My main concern is that much of this information is redundant, it is simply scattered across the internet right now and needs to be centralized and augmented with contact info for the relevant researchers/clinicians.  Can algorithms be devised to eliminate much of the effort for you?  If so, how to possibly do it all, disease by disease and mutation by mutation?  OMIM has kind of done this although it is written by scientists for scientists, with no possibility for interaction.  Will this be directed towards patients?  If so, can it be built on top of something like OMIM?  And how to exclude people with personal genomes/exomes that have polymorphisms which are not causative for a given gene from clogging up the system?

One of the points of doing this is for the very purpose of developing and maintaining a portal for information that is currently scattered across the internet and to reduce any redundancy by tying it all together. The goal is to improve knowledge and information transfer by making it easier to access. This will be accomplished in part by maintaining an updated portal. An important part of this, of course, is also maintaining links to other sources of user generated content (e.g. OMIM, WikiGenes). Part of this effort will also be to update content on other sites as the need is identified. If there are algorithms that can automate this, we are happy to employ them - although, we presume the initial set up and curation would fall to the portal hosts. (please see the next response regarding your point of scalability).

A centralized information management service (IMS) would be a useful service for patients and physicians.  It would be useful to know what the frequency of symptomatic mutations are in this gene to get an idea of the clinical relevance of identifying allelic variants.  If the frequency of allelic variants in an asymptomatic population is low, and clinical presentation is common for any one manifestation then screening larger populations may be useful, and informing patients with allelic variants may have value.  It is difficult to estimate the value of an IMS in the absence of more information on allelic variant frequency and penetration.

Thank you for the comment Sigurd, we agree. Presently, this information is available but without being in a centrally located and curated location. For example, we recently sequenced 96 patients with intracerebral hemorrhages for both COL4A1 and COL4A2 but these data are found in supplemental data of the manuscript that even collaborators overlook. This type of information can all be easily accessible in one web interface using the format of the Leiden Open Variant Database (LOVD) as our platform.

Commenting is closed.