High-Throughput Digital Phenotyping

Data collection and analysis platforms for Android and iOS devices

From Data to Analytics to Insights


Beiwe is our open source data collection platform for Android and iOS devices. It consists of two native smartphone applications and an AWS-based system back-end. Its development was enabled by an NIH Director's New Innovator Award to Dr. JP Onnela in 2013.


Forest is our open source data analysis platform for Beiwe data. This Python library can be run locally, but it also integrates directly with the Beiwe back-end on AWS to provide scalable on-demand analytics. Its Tableau API supports customizable workbooks and dashboards.


Our research focus is the development of mathematical and statistical methods for intensive high-dimensional data. We also participate in various applied digital phenotyping studies in medicine and public health with emphasis on central nervous system disorders.

Our Approach

Social, behavioral, and cognitive phenotypes have traditionally been difficult to study due to their temporal and contextual dependence. The standard approach to learn about, say, behavior is to administer surveys, which are cross-sectional, subjective, and often burdensome to subjects. We believe that the ubiquity and capability of smartphones–when coupled with appropriate data analytic techniques–can be part of the solution. We have coined the term Digital Phenotyping to refer to the “moment-by-moment quantification of the individual-level human phenotype in situ using data from personal digital devices, in particular smartphones.” Unlike the large majority of commercially available smartphone applications, Beiwe is intended for use in research. Most applications, out of convenience, use software development kits (SDKs) made by Apple and Google which usually generate unvalidated behavioral summary measures using closed proprietary algorithms that do not meet the high standards of reproducible science. In contrast, Beiwe collects raw sensor and phone use data and Forest analyzes these raw data. That means that data collection and data analysis are customized to address specific scientific questions of interest rather than modifying scientific questions based on what data happens to be available.

Commitment to Reproducible Science

We focus on creating tools that enable researchers to conduct studies that are truly reproducible. Unlike the large majority of commercially available smartphone applications, Beiwe is intended for use in research. Its companion library, Forest, is intended for analyzing data collected with Beiwe. Both are open-source software, and they have been published under the permissive 3-clause BSD (BSD-3) license. Every single Beiwe data collection setting is captured as a JSON formatted study configuration file which can be easily imported to or exported from Beiwe. Public sharing of study configuration files makes it easy for anyone to reproduce a study “as is.”

Using Beiwe and Forest


The Onnela Lab has limited opportunities for direct research collaborations using the Beiwe platform. In this model, the Onnela Lab provides the platform for the study and collaborates directly with you to advance the research project. These collaborations are usually funded by a joint grant or by an industry partner.

Beiwe Service Center

This option makes Beiwe available to any academic or commercial entity under the software-as-a-service model. As a formal service core of Harvard University, its pricing is based on the extent of platform use. This is not a scientific collaboration with the Onnela Lab but an effort to make it easy for others to use Beiwe in their research.

Open Source

Beiwe and Forest are both open source software under the BSD-3 license and as such are available at no cost to investigators worldwide. Under this model, individuals or institutions can deploy their own instance of Beiwe and Forest on AWS. Investigators are responsible for all expenses and software is provided as is.

Get Started with Beiwe

Beiwe Service Center

This software-as-a-service model makes it easy to use Beiwe. To determine pricing, we just need to know the number of subjects, study duration, and average length of follow-up per subject. Harvard University also requires us to have a copy of your study IRB. Once we have that, we can usually create your study within 48 hours.