2024 Harvard Data Science Initiative Research Fund Request for Proposals

Track A is designed to coalesce and accelerate methodologically-focused research across disciplines. We are especially interested in projects that intersect with or are likely to have impact within or across the DSI’s research themes: 

1. Data-Driven Scientific Discovery (includes discovery of new materials, drug and gene discovery, astronomy, neuroscience, environment and health including climate change mitigation/resilience, greenhouse gases, agriculture/crop science, food security/sustainability, air pollution, water scarcity/contamination, biodiversity loss, deforestation, marine pollution and fishery depletion) 
2. Markets and Networks (includes networks and influence, innovation and crowds, digital economy, labor economics, data-driven decisions, blockchain)
3. Health and Biomedical Science (includes precision medicine, drug repurposing, public health, medical informatics, diagnostics, personal devices)
4. Evidence-Based Policy (includes equality of opportunity, healthcare economics, democracy and governance)

We also welcome work that is primarily methodological or addresses deficiencies in data sets, including gaps or remediating poor quality data. We are interested in promoting advances across many areas that relate to the science of data, including causal inference,    visualization, generative AI, scalable and robust inference, experimental design, interpretability and robustness, ethics (including privacy and fairness), control of false discovery, human-in-the-loop, reinforcement learning, multi-agent systems, adaptive data systems, deep learning, theoretical foundations, reproducibility, and data sharing.

Proposals submitted via Track B should seek to illuminate the varied factors that currently impede trusting relations between the producers and users of scientific information. Projects should leverage data science to analyze the breakdowns in public trust, and to ask what steps could be taken to promote better mutual understanding. Examples of topics that may be of interest include: 

• The processes and products of data science be made more transparent, and how might strategies of democratization affect the trustworthiness of science?
• Explorations into the juxtaposition of and tension between individual or group autonomy and trust in science  
• Methods of visualizing data and how they affect the ways that different groups assess the trustworthiness of that data
• The role of team structures in science in the trustworthiness of their results
• Propagation of conspiracy theories
• Public policy influencing through global data citizenry and communication
• Misinformation and how data and public health relate
• Science advocacy, transparency, and ethics
• Improving AI integrity, credibility, and adoption
• Optimizing the value and credibility of AI deployment, particularly in the clinical realm (e.g., clinical diagnosis and predictive modeling, health monitoring/analytics, and clinical decision-making)
• COVID-19: data science-driven learnings for the next pandemicThese examples are illustrative only; applicants are encouraged to think more broadly, and we welcome alternative research directions.  

This track is reserved for previous applicants to the Trust in Science fund ONLY.

The goal of Track C is to promote and support cross-disciplinary collaboration between data scientists at the postdoctoral level. We are particularly interested in funding research proposals that aim to: 

1. Investigate novel applications of data analysis techniques (broadly defined), particularly by transferring methodology from one field to another; 
2. Share and combine existing but distinct data sets to gain new insights into a problem; 
3. Involve new collaborations between researchers in separate departments (or fields of research); 
4. Explore new methods that may help to improve the public understanding of complex technical issues or areas of research.

Principal investigator and collaborator information. Include affiliations and professional titles

The summary should be written in layperson’s language and may be used by the HDSI in promotional materials.

Describe the research plan, which should be comprehensible and accessible to non-specialists. The project description should include the following:

Description of proposed research:

1. Statement of real-world problem the project addresses and significance
2. Goals and proposed research questions
3. Experimental plan/approach
4. The potential impact of the proposed work

Data and computational needs*:

1. Description of any computational services that are needed (e.g., high performance cloud computing, cloud-based data storage, coding, data research platforms, database architecture, software engineering, analytics, visualization, etc.)
2. The role of data science and computational analysis 
3. The role of any specialized technical support including, for example, a data engineer or professional data scientist 
4. Datasets required for the study: description, availability, validation, quality

Project Management: 

1. Milestones/phases
2. Go/no-go decision points (i.e., stage gates)
3. Contingency plans/Pre-mortem if the research does not go as planned 
4. Anticipated outputs and measures of success

Abridged CV or biosketch (limit two pages) for each PI and co-PI.

List of all current or pending external sources of grant support for the proposed project, and information on any external funding you have applied for or intend to apply for to support the project. 

Budgets should provide enough information to convey the alignment of costs with the project and use of funds. Faculty are encouraged to work with their grants administrators when including personnel and fringe benefits. School assessments and/or indirect costs should not be included in your budget (the DSI will be covering any assessment above the award amount).

Examples of eligible expenses include:

• Personnel, such as postdocs, research staff, undergraduate students
• Travel (domestic and international)
• Acquisition of datasets
• Publishing costs

The following expenses are NOT eligible for funding:

• Faculty salary (for all tracks) and postdoc salaries for (track C only)
• Professional development and education
• Subcontracts outside Harvard
• Equipment