HOPE Consortium
Healthy Outcomes of Pregnancy for Everyone through Science, Partnership, and Equity
The HOPE Accelerator is a data collection and storage platform that allows consortium partners to collect and share data seamlessly and securely.
Partners are able to collect and combine different kinds of data (clinical, molecular, survey, geographical) from one or multiple studies. The platform is designed to boost progress by offering a place where data from multiple studies can be combined and analyzed safely and securely. No identifying information for participants is kept within the shared system, which insures that information is shared without any compromise to privacy.
Resources Include:
- Combined birth, death, and hospital records for more than 10 million California births with linkage from one year prior to one year after birth
- For a subset of these mother-baby pairs, we have linked prenatal and newborn screening results
- Data from the California 1000 study
- Combined birth and hospital records for 500 pregnancies with and without preterm birth. Includes robust multi-omic testing results (metabolomics, immune markers, full genome sequencing, lipids, heavy metals, cortisol)
- Data from the HOPE COVID-19 study including for more than 400 mother-baby pairs from around the world enrolled in 2020-2024 with infant follow-up to 18-months
- Robust qualitative and quantitative survey data
- Activity, blood pressure, and biospecimen collection and testing for a San Francisco-based subset
Coming (these data are currently being collected and analyzed by partners within the collaborative and are expected to be made available across the collaborative in 2025-2026)
Data from the PROMPT/PREMO study being collected as part of a multi-site R01 (with Indiana University and the University of Iowa)
- Includes sequential metabolic testing of newborns with preterm birth < 32 followed to discharge as well as extensive hospital-based data collection for mothers and babies
- Extensive specimen collection with microbiome testing in a subset as well as brain imaging
Data from the THRIVE study being collected as part of a funded RCT focused on assessing the impact of digital cognitive behavioral therapy on anxiety in low-income pregnant people
Combining Data
Through the HOPE Accelerator, investigators can combine (pool) data from multiple studies to look for common patterns and trends. We do this by establishing an initial data footprint (data type, format, units) from the core study. Then, subsequent data is tested and translated to fit into that initial footprint.
Once completed, investigators are able to look at patterns across multiple studies using either a dashboard, standardized reports, by using Python or R in an embedded notebook environment or a 3D data explorer.
Data and Standardized Reports
For large studies like the HOPE COVID-19 study, a study dashboard can be created in the HOPE Accelerator that will allow the investigator and study team to track the number of participants in the study overall and by grouping.
Data can also be translated into standardized reports that capture data that the study team wants to be able to track and highlight on an ongoing basis.
2D Analyses Using Notebook
The HOPE Accelerator includes embedded coding languges like Python and R. This capability allows investigators to carry out all analyses that these coding languages allow and visualize the results within the platform.
The HOPE Accelerator also allows for supervised and unsupervised machine learning and artificial intelligence, which allows investigators to look for patterns in large sets of complex data.
3D Exploration
Using the HOPE Accelerator, investigators are able to explore relationships between multiple types of data, across multiple participants, and across time.
The 3D tool tags clinical, laboratory and molecular data and then visualizes that information.
Investigators are able to drive through the 3D tool using an embedded graphics interface. This capability allows investigators to identify potential relationships and then explore them using the embedded network tools. This allows for rapid discovery.
Looking at Data Accross Time
Using the HOPE Accelerator and the 3D explorer, investigators are able to look at participant data over time. Looking at longitudinal data through a 3D perspective allows patterns to be quickly identified and explored.
Investigators are able to identify potentially important patterns in minutes rather than in the days and weeks that identification of promising targets would take using only analysis tools.
Exploring Geographic Patterns
Using the HOPE Accelerator platform, investigators are able to identify signals in the data that may be specific to certain geographies.
Participant data can be overlaid onto maps at a neighborhood, city, state, or country level.
The visualization tool provides clues that can be further investigated in the notebook environment or using other platforms like ArcGIS via the use of data download.
