MatchMiner: An Open-source Computational platform for Clinical Trials Genomic Matching

The challenge of matching precipitant (patients) for cancer trials is not easy nor simple. The current methods for patient recruitment for clinical trials result in failure [5]. Our topic of the day MatchMiner is designed to help researchers overcome this challenge.

MatchMiner is an open-source computational platform with a specific focus on patient genomic profiles to precision cancer medicine clinical trials. It is intended for researchers with software development skills.

The project is developed by researchers at Dana-Farber Cancer Institute (DFCI) which is known as a comprehensive cancer treatment and research institution in Boston, Massachusetts, United States. It's also an affiliate of Harvard Medical School.

What is Clinical Trial Markup Language (CTML)?

The Clinical Trial Markup Language or (CTML) is a new standard that aims to make a new standard for structured clinical trial data.  

It is introduced by Ethan Siegel in his article "Making Clinical Trial Eligibility Machine Readable". Siegel is also one of the core researchers who developed MatchMiner.

Here are two examples captured from Siegel's article:

How does CTML look like? (src. Making Clinical Trial Eligibility Machine Readable)
How does CTML look like? (src. Making Clinical Trial Eligibility Machine Readable)

MatchMiner is formed of 3 parts:

  1. MatchMiner engine
    The engine is responsible for parsing structured clinical trial data. It follows the Clinical Trial Markup Language (CTML) Standard.
  2. MatchMiner API
    The API is built as an interface for the engine. It is built with Python3.6, uses Eve framework and MongoDB a NoSQL database engine for data storage.
  3. MatchMiner UI
    The MatchMiner UI is a front-end application that is built on top of the API. It's written in JavaScript (NodeJS 10.16.9) and uses AngularJS. It uses Gulp for building and deployment.

    MatchMiner UI offers user authentication, user roles management, and two modes: a clinical trial investigator mode, and oncologist mode.

How does MatchMiner work?

It uses patient records extracted from the electronic medical record (EMR) or electronic health record (EHR) systems and clinical trial data in Clinical Trial Markup Language standard.

The MatchMiner platform matches patient-specific genomic events to clinical trials, and makes the results available to trial investigators and clinicians via a web-based platform.
MatchMiner's mode (src, MatchMiner)
MatchMiner's mode (src, MatchMiner)

Data collection and trial matches with MatchMiner (src, MatchMiner)
Data collection and trial matches with MatchMiner (src, MatchMiner)

Features

  1. Open-source
  2. Complete API support
  3. Developer-friendly
  4. Extensible though plugins and extensions
  5. Well written documentation
  6. Command-line interface
  7. Elasticsearch support

Platforms

MatchMiner can be installed in any platform that has Python and NodeJS.

Technologies used

  • Python3.6
  • NodeJS
  • Eve Rest Framework
  • MongoDB
  • Angular Framework
  • Gulp
  • Docker

License

MatchMiner is released as an open-source project under Apache License 2.0.

Apache License 2.0 (src. GitHub)

Conclusion

Development-wise, MatchMiner follows a good design pattern to ensure a stable and fixable product. Splitting the engine from the user-interface is a good decision as well as proving a headless REST-API on top of the engine which allows developers to integrate it easily in their projects, or build their own applications on top of it.

It's an approach we rarely see from developers with research background as many tend to use old-school approaches to solve their research computational problems.

Resources

  1. Official Website: https://matchminer.org/
  2. MatchMiner: Documentation
  3. https://github.com/dfci/matchminer
  4. https://www.dana-farber.org/
  5. https://medium.com/matchminer/making-clinical-trial-eligibility-machine-readable-3fd3dffb9a8b
  6. https://medium.com/matchminer/clinical-trial-matching-at-the-dana-farber-cancer-institute-f4c57ac18ab0
  7. https://cancerres.aacrjournals.org/content/80/16_Supplement/3382
  8. https://blog.dana-farber.org/insight/2017/01/how-do-i-find-the-right-clinical-trial-for-me/
  9. https://www.hbs.edu/news/releases/Pages/matchmaker-wins-hbs-kraft-challenge.aspx

Photo by Anna Shvets from Pexels



  • TiddlyWiki is a lightweight, portable multipurpose note-taking app. Created by Jeremy Ruston and powered by a strong community, TiddlyWiki became  productivity tool for many. Because it is highly customizable and portable, you can use it for almost anything. Besides, It has a growing ecosystem filled by its community with plugins,...Read more...

  • SageMath is a free open-source mathematic software for mathematicians, data scientists and statisticians. It is built on top of many mathematic python packages.   SageMath features include animated graphs, interactive plots,  portable version that works directly from USB stick, interactive Python interface, notebook, rich documentation and more. SageMath is an ideal...Read more...

  • Traditionally, a research proposal is what a student submits to their academic advisor before working on their thesis. However, colleges in the USA, UK, and other English-speaking countries also have research proposals for smaller research assignments. Basically, it’s a paper that describes the future research that the student would...Read more...

  • Development of Soft skills in education and work With the development of the world and innovations in it, we also need to change. You can find any information on the internet, watch videos about space, other dimensions, write help me edit my essay, and you will get an answer. As...Read more...

  • Managing clinical trials is not that just a CRUD operation. It's a process which involve patients, patient records, clinical trials records and  maintain the patient privacy as well as keeping the data secure. Phoenix CTMS is one of a kind open-source project, It's built to help researchers and teams for...Read more...