Skip to content
Atlas of Computational Cell Reprogramming

Companion review

Computational Cell Reprogramming: Shared Principles, Distinct Constraints

The paper teaches the taxonomy; this site applies it. Together they form one artifact: the review introduces the four-level classification and the audit; the Atlas tracks the field continuously as new methods appear.

The thesis

Across the four reprogramming modalities, computational methods address one inverse-design objective:

Methods differ less in what they ask than in how faithfully they answer it. The four-level taxonomy ranks them on exactly that axis.

Sections at a glance

  1. §1

    Introduction

    Why computational cell reprogramming is timely now: single-cell multiomics, perturbation atlases at hundreds of millions of cells, and the widening scope of reprogramming applications across regenerative medicine, disease modeling, drug discovery, and immunotherapy.

  2. §2

    A shared intervention-design problem

    iPSC induction, directed differentiation, transdifferentiation, and phenotype reprogramming can all be written as constrained variants of the same inverse-design problem; modalities differ in target geometry, admissible interventions, timescale, and feasibility penalty.

  3. §3

    Four levels of inverse-design fidelity

    L0 target-feature discovery → L1 network-informed heuristic → L2 proxy inverse design → L3 explicit model-based inverse intervention. The taxonomy is the paper's main conceptual contribution and the spine of this Atlas.

  4. §4

    Where the unification strains

    Seven biological nuances grouped into representation, controllability, and validation families. The shared algorithmic syntax holds; the biological objects occupying it change across modalities.

  5. §5

    The infrastructure gap

    A reproducibility and validation-data audit indexed by inverse-design level. Code availability ≠ rerunnability; FAIR4RS compliance is a near-floor problem across every level; validation data is almost entirely siloed.

  6. §6

    Outlook

    Shared problem statements, DREAM-style community benchmarks, reproducibility reframed as reusable scientific infrastructure, and multiscale predictive modeling as the operational program for the next stage of the field.

Citation

Vera-Licona P. Computational Cell Reprogramming: Shared Principles, Distinct Constraints. Under submission, 2026.

The full PDF, BibTeX entry, supplementary tables, and reproducibility audit data will be made available here at publication.