The Defense Advanced Research Projects Agency (DARPA), under the Defense Sciences Office (DSO), has released the full Program Solicitation for Protein ENGineering (PENG), DARPA-PS-26-129. The solicitation moves PENG from concept and Proposers Day into an active, multi‑phase program seeking innovative proposals to build a multifunctional, target‑agnostic platform for precise, programmable editing of endogenous proteins.
PENG’s core objective is to transform proteome engineering from a collection of bespoke tools into a universal, software‑defined discipline. DARPA is asking teams to integrate programmable targeting, generalizable effector chemistries, and multiplexed control into a single platform that can edit mature proteins in situ and deliver transient, mechanism‑based therapeutic outcomes without permanent genomic modification or traditional pharmacology.
Technical Pillars and Phases
The solicitation frames PENG around three pillar capabilities and a 42‑month, three‑phase structure.
Core Pillar Capabilities
Performers must build an integrated pipeline that combines:
- Programmable targeting using generative deep learning and structural computational chemistry to design de novo binders that can dock editing machinery to virtually any epitope on a mature, native protein.
Generalizable chemistry that moves beyond single‑use point solutions toward reusable mechanisms such as programmable intein chassis, logic‑gated chemistry, or co‑translational editing at the ribosomal exit tunnel.
Multiplexing capability that deploys multiple, orthogonal, non‑interfering editing modules simultaneously, achieving multi‑site reprogramming in real time without destabilizing protein structure or inducing toxicity.
All work must focus on direct editing of endogenous proteins in native cellular environments, with a credible path from single‑cell models to higher‑order tissue complexity such as 3D cell layers or organoids. Human subjects research is out of scope; animal work is allowed, particularly in Phase 2.
Program Phases
PENG is structured as a 42‑month effort with multiple anticipated awards under Other Transaction Agreements for Prototype:
- Phase 1A – Platform Foundations (Months 1–18)
Teams establish the foundational biochemistry and demonstrate target generalizability using self‑selected targets, with precise, quantitative functional outcomes. Platforms must successfully edit at least three distinct structural protein families (for example, globular, transmembrane, fibrous) to prove generalizable chemistry. - Phase 1B – Capability Demonstration (Months 19–24)
A six‑month option period that serves as a gate. Performers are evaluated head‑to‑head on difficult edits against DARPA‑selected, previously unannounced protein targets under short‑notice constraints in complex tissue systems. This phase is explicitly designed to test programmability and adaptability; selection into Phase 2 depends on Phase 1B performance. - Phase 2 – Advanced System Integration (Months 25–42)
The final 18‑month phase pushes validated platforms into higher‑order biological systems, transitioning from tissue models into more complex cellular systems and optimizing kinetics, safety, delivery, and multiplexing at scale. Phase 2 can include animal model work to demonstrate context‑dependent, reversible control over physiological functions.
Across phases, DARPA has defined quantitative program metrics for targeting efficiency, edit site diversity and reuse, multiplexing across cell types, and stability/readiness for future transition, and proposals must explicitly address these metrics with theory, modeling, and planned experimental validation.
Key Dates
The solicitation sets an aggressive near-term schedule, with all times in Eastern Time:
- Posting date: August 7, 2026
- Proposers Day: August 10, 2026
- Questions due: August 17, 2026, by 4:00 p.m.
- Proposal abstracts due: August 24, 2026, by 4:00 p.m.
- Tentative full proposal due date: October 19, 2026, by 4:00 p.m.
DARPA anticipates multiple awards under Other Transaction Agreements for Prototype, with a total period of performance of 18 months for Phase 1A, 6 months for Phase 1B, and 18 months for Phase 2.
Why PENG Matters
With the release of the official solicitation, PENG moves from concept to execution on a bold vision: a generalizable “CRISPR for proteins” that operates at the functional execution layer rather than the genome.
DARPA’s framing highlights several strategic shifts:
Direct, transient control of mature proteins as a route to dynamic biological resilience, allowing physiological responses to be scaled up or down just long enough to manage acute stress, toxicity, injury, or pathogenesis.
Use of generative AI and synthetic biology to overcome historic biophysical barriers to accessing folded domains and performing multiplex edits without misfolding.
A platform mindset, where successful teams deliver reusable, programmable systems for proteome engineering, not one‑off tools tied to single targets.
For teams in protein engineering, synthetic biology, structural biology, and advanced in vitro or organoid systems, PENG is a significant opportunity to help define what programmable protein editing looks like as a discipline. It also aligns with broader defense and biotech priorities around tunable mechanism‑based therapeutics, advanced materials, and sustainable solutions driven by precise control of protein function.
EverGlade is a national advisory firm helping innovators navigate the federal funding ecosystem. We support companies across the funding lifecycle, from early-stage strategy through proposal development, negotiations, and post-award execution, ensuring you win the award and deliver the program.
For additional information on where your capabilities could plug into this program, schedule a conversation with our team.
