Reproducible Cells and Organoids via Directed- Differentiation Encoding | 21 608

FAQs: NSF RECODE (Reproducible Cells and Organoids via Directed-Differentiation Encoding), Funding Opportunity Number 21-608

What is the main goal of the NSF RECODE opportunity (21-608)?

The goal is to make cell differentiation far more predictable, controllable, and reproducible by moving beyond trial-and-error protocols and developing robust, validated design rules and mechanistic models that can reliably guide undifferentiated cells into mature, functional cell types or organoids.

What does NSF mean by treating differentiation as a "dynamic, information-rich process"?

RECODE expects projects to monitor differentiation in real time, steer it using feedback control, and verify outcomes using rigorous quality assurance methods, so that the end-to-end differentiation strategy can be reproduced by others rather than being dependent on lab-specific know-how.

What kinds of research outcomes is RECODE trying to produce?

RECODE is aiming for practical, end-to-end differentiation strategies that others can reproduce, supported by generalizable design rules, mechanistic and predictive models, real-time sensing, feedback control, and quality assurance that together enable reliable production of reproducible cells or organoids.

Is RECODE focused on making a specific cell type or organoid?

No. The program is deliberately broad. Teams may choose any undifferentiated cell type and aim for any mature functional cell type, tissue-like assembly, or 3D organoid with real-world relevance. The emphasis is on building transferable rules and control strategies, not just achieving a single end product once.

What biological systems and application areas are considered relevant?

Targets can span many physiological systems, including cardiovascular, nervous, and immune systems, among others. NSF also signals interest beyond medical applications, including biotechnology, materials, sensing, and environmental domains.

Are non-model organisms or non-standard systems allowed or encouraged?

Yes. NSF encourages the use of non-model systems (for example, systems other than standard mouse models), reflecting the program's broad scope and interest in generalizable principles.

Can projects use human cells?

Yes. Investigators may choose any undifferentiated cell type from any animal species, including human cells.

What makes a project "responsive" to the RECODE solicitation?

Responsive projects combine engineering-style control with deep biological understanding to uncover mechanisms and produce validated, generalizable differentiation rules and control strategies. Projects should treat differentiation as something that can be measured, modeled, steered, and verified for reproducibility.

What kinds of scientific and engineering elements does NSF expect proposals to include?

The solicitation emphasizes: mechanistic understanding of differentiation decisions; the combined effects of signaling molecules and pathways, gene regulatory networks, receptors and promoters, molecular markers, and chemical and mechanical conditions; the importance of timing and dynamics; real-time monitoring; feedback control; and rigorous quality assurance to ensure reproducible outcomes.

How important is the cellular environment in RECODE projects?

It is central. The opportunity emphasizes cell-cell interactions, cell-microbe interactions, and cell-extracellular matrix (ECM) interactions, and expects teams to use mechanistic insight about these interactions to intentionally manipulate developmental trajectories, not merely observe them.

Does the program expect teams to explain "how" differentiation happens, not just "what factors" are used?

Yes. RECODE calls for uncovering underlying mechanisms and molecular machinery that drive development, including how timing and dynamics shape outcomes, rather than only listing ingredients or cues.

Is real-time sensing and feedback control required or just encouraged?

The program is explicitly framed around treating differentiation as measurable and steerable, including real-time monitoring and feedback control. Proposals are expected to align with that vision as part of delivering reproducible, controllable differentiation strategies.

What role do "design rules" and mechanistic models play in RECODE?

They are core deliverables. NSF is looking for robust, validated, generalizable rules and mechanistic, predictive models that can reliably guide differentiation along predetermined paths and help track and adjust trajectories as cells change.

How does RECODE relate to the U.S. bioeconomy and biotechnology capabilities?

RECODE is positioned as part of NSF's broader push to build U.S. biotechnology capabilities supporting the national bioeconomy by enabling more reliable and scalable differentiation, which can underpin biomanufacturing, biomaterials, and reproducible organoid production for uses like drug testing and other applications.

Does RECODE include topics related to immunity or host-microbe relationships?

Yes. The solicitation highlights fundamental questions relevant to immunity and host-microbe relationships, such as how systems move between symbiosis and disease and how immune responses emerge during infection.

What is the expected team structure for RECODE proposals?

Team structure and integration are central requirements. Proposals must be submitted by teams of three or more PI/co-PIs and senior personnel with complementary expertise. A single PI, or a PI plus only one additional senior person, is not allowed and will be returned without review.

Why does NSF emphasize convergence and integration in this program?

Because RECODE is designed around convergence research. The solicitation expects genuine integration across disciplines to address differentiation as a unified, end-to-end controllable process rather than tackling only one isolated slice of the problem.

What disciplines and expertise areas are explicitly relevant to RECODE?

NSF encourages interdisciplinary teams spanning areas such as engineering, computation, sensing, systems and synthetic biology, developmental biology, stem cell biology, mechanobiology, cell physiology, microbiology, immunology, and biophysics.

What kinds of projects are considered out of scope or non-responsive?

NSF indicates proposals may be deemed non-responsive if they focus on only one slice of the differentiation problem, mainly aim to produce an engineered living product without advancing underlying understanding and controllability, sit clearly inside a single traditional NSF core program area, focus on isolated stages or mechanisms without building unified differentiation rules and control strategies, or center purely on building engineered cells, tissues, organ-on-a-chip systems, or organoids without developing broadly useful differentiation rules.

Does RECODE fund work that only builds an organoid or engineered tissue?

Work centered purely on building engineered cells, tissues, organ-on-a-chip systems, or organoids is described as not responsive if it does not develop broadly useful differentiation rules and strategies for controllability and reproducibility.

Are preliminary proposals required?

Yes. Preliminary proposals are mandatory before a full proposal can be submitted, reflecting NSF's expectation that teams will use the preliminary stage to demonstrate cohesive planning and true integration across disciplines.

What is the maximum project duration and budget described for RECODE?

RECODE supports collaborative grants with project durations up to 4 years and budgets up to $1,500,000 total, with the expectation that the budget matches the scope and need for complementary expertise.

How many awards does NSF anticipate making under this opportunity?

The opportunity record states NSF anticipates making around 7 awards.

Why might the award ceiling appear as 0 in summary data if the solicitation mentions up to $1.5M?

The summary data lists an award ceiling of 0 even though the solicitation text specifies budgets up to $1.5M. This typically indicates the cap is being managed through the solicitation guidance rather than the system field.

What is the key difference between trial-and-error differentiation and what RECODE is asking for?

RECODE is asking for validated, mechanistic, and controllable approaches with real-time monitoring, feedback steering, and quality assurance so differentiation can be guided along predetermined paths and reproduced across settings, rather than relying on ad hoc protocols that may not transfer well between labs.

What does "quality assurance" mean in the context of RECODE?

Within the solicitation's framing, quality assurance refers to rigorous methods to verify that differentiation outcomes meet defined standards and are reproducible, supporting transparent and transferable end-to-end differentiation strategies.