New manuscript, Megathon event, and project updates

Hi and many apologies for the radio silence!

We’ve been waiting with bated breath to hear back on a few submissions.

First, our manuscript submission for the Automated Knowledge Base Construction has completed the anonymous peer review process and has been accepted for the conference. We’ve uploaded the accepted submission to BioRxiv.

The manuscript along with the data and code used to generate the figures in the manuscript can be found at:

Most Important--Contributors to the data set used who gave us their permission to acknowledge them: https://mark2cure.org/blog/relation-paper-contributors/

Manuscript: https://www.biorxiv.org/content/10.1101/564187v1

Data: https://github.com/gtsueng/M2Crelnb/tree/master/data

Analysis code: https://github.com/gtsueng/M2Crelnb

Here are some key findings from the analysis of the data from the Relationship Extraction Module:

• Good relationship extraction is difficult without good Concept Recognition. When it comes to doing concept recognition, we often get asked the question, “Can’t machines do that?” And, as those of you who’ve been contributing to Mark2Cure know--yes, but there’s definitely room for improvement. A large part of error in the Relationship Extraction Module stems from concept recognition errors.

• Design issues also hamper user contributions. We’ve known all along that users are good, but that design issues can muck things up for them. Still, it can’t be stated to the scientific community enough! Data issues in citizen science are usually design issues.

• The relationships obtained via Mark2Cure are very different from those found using an algorithm, and we think that these two methods can complement one another nicely to cover gaps within each system.

The other submission we were waiting on was our funding application to combine Mark2Cure and active machine learning to identify diseases based on their clinical phenotypes. Unfortunately, this was not approved for funding. We are still investigating alternative applications for Mark2Cure in order to further its development and will let you know when we make progress on that end.

Rare Disease Day

Although we had high hopes for pushing this newsletter out on Rare Disease Day (yesterday) in order to acknowledge and thank our contributors from the rare disease community, we were unfortunately just not fast enough. Although we missed sending this out on Rare Disease Day, rare disease is not something that should be acknowledged just once a year. Rare disease patients and family live with the disease every day, and we are grateful to the rare disease folks in our community who have taken the time to share their stories with us, contribute to our efforts, and inspire researchers like us. We know that research on your disease is probably moving much more slowly than you’d like, and hope that in the future we’ll be able to speed things up (though we’re also pushing forward slowly right now.)

Some news from our friends about an upcoming event for Alzheimer’s Disease Citizen Science

StallCatchers, the citizen science project from our CitSciMedBlitz event last year, is aiming for a Guinness World Record and would love for you to take part in this effort. The goals is to surpass the world record for ‘largest practical science lesson (multiple venues)’ which currently held at 13,701 participants. If you’re local to San Diego, the San Diego Public Library (in particular, the La Jolla branch, whom we’ve partnered with for the San Diego Citizen Science Day Expo) is one of 9 cities in the U.S. where the libraries are engaged in the planning, preparation, and evaluation of this event.

If you’re interested in helping researchers to understand some of the circulatory underpinnings of Alzheimer’s disease, you can take part in this StallCatchers Megathon that happens between 10.30 am and 12.30pm on Saturday April 13th.

How you can take part:

• Sign up for at stallcatchers.com (https://stallcatchers.com/main#)

• If you’re in San Diego, make sure you and your friends join our online team after they sign up ('SD library's Finest CITYzens') so that the participation is credited to San Diego.

• Promote it in your community.

• Host a group event on April 13th.

• Use the project as a springboard for building community interest in other neighborhood science ideas of your own.

• Any age can take part. The project involves looking at scans of mice brains.