1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
|
# Cell-specific expression
Use [[gene editing|gene-editing]] technology to make a (re)programmable system for cell-type-specific expression of genetic payloads for [[gene therapy|gene-therapy]].
* [An enhancer-AAV toolbox to target and manipulate distinct interneuron subtypes](https://www.cell.com/neuron/fulltext/S0896-6273%2825%2900349-6)
* [A suite of enhancer AAVs and transgenic mouse lines for genetic access to cortical cell types](https://www.cell.com/cell/fulltext/S0092-8674(25)00513-6)
* <a href="https://www.cell.com/cell-reports/fulltext/S2211-1247(25)00501-7">Enhancer AAVs for targeting spinal motor neurons and descending motor pathways in rodents and macaque</a>
* [Combining machine learning and multiplexed in situ profiling to engineer cell type and behavioral specificity](https://www.biorxiv.org/content/10.1101/2025.06.20.660790.abstract)
* <a href="https://www.cell.com/cell-reports/fulltext/S2211-1247(21)00067-X?elqTrackId=d63aee9419524ab9b8e1a8841e45579a">Functional enhancer elements drive subclass-selective expression from mouse to primate neocortex</a>
* [Advances in AAV technology for delivering genetically encoded cargo to the nonhuman primate nervous system](https://www.sciencedirect.com/science/article/pii/S2665945X23000141)
* <a href="https://www.cell.com/neuron/fulltext/S0896-6273(25)00313-7">Machine learning identification of enhancers in the rhesus macaque genome</a>
* <a href="https://www.cell.com/cell-reports/fulltext/S2211-1247(25)00808-3">RNA-programmable cell-type monitoring and manipulation in the human cortex with CellREADR</a>
* <a href="https://www.science.org/doi/abs/10.1126/science.adp3957">Enhancer-driven cell type comparison reveals similarities between the mammalian and bird pallium</a> (2025)
* <a href="https://www.cell.com/cell-genomics/fulltext/S2666-979X(25)00135-1">Evaluating methods for the prediction of cell-type-specific enhancers in the mammalian cortex</a> (2025)
* [Predicting gene expression from DNA sequence using deep learning models](https://www.nature.com/articles/s41576-025-00841-2) (2025)
# Practical and scalable multi-sensor cell-specific expression techniques
**1) miRNA signature logic (“who am I?”) → CRISPR gate**
* **What:** Read endogenous miRNA patterns with **miR-OFF** (detarget a transgene where a miRNA is high) and **miR-ON** (activate where it’s high), then drive a CRISPR layer.
* **How to wire:**
* Put **miRNA target sites** (e.g., miR-122 for liver, miR-142-3p for hematopoietic) in the 3′UTR of (i) your payload, (ii) **anti-CRISPR** proteins (Acr), or (iii) dCas9/base editor halves to implement **AND/NOT** gates. This is small-payload friendly (fits in AAV). [PMC](https://pmc.ncbi.nlm.nih.gov/articles/PMC4873559/?utm_source=chatgpt.com), [ScienceDirect](https://www.sciencedirect.com/science/article/pii/S1525001616304014?utm_source=chatgpt.com)
* **Why good:** Extremely portable, widely used in vivo (AAV, mRNA). **Cas9-ON via miRNA-repressed Acr** is a great general switch (low background, strong selectivity). [PMC](https://pmc.ncbi.nlm.nih.gov/articles/PMC6648350/?utm_source=chatgpt.com)
**2) gRNA-programmable promoters → deep/wide logic without new proteins**
* **What:** Use **CRISPRa/CRISPRi-responsive synthetic promoters/operators** (promoters studded with gRNA target sites) controlled by dCas9-VPR/SAM or KRAB. You can stack many orthogonal promoters and gRNAs to implement multi-input logic with just a handful of proteins. [Nature](https://www.nature.com/articles/s41467-022-33287-9?utm_source=chatgpt.com)
* **Why good:** Very **reprogrammable** (swap gRNAs, not proteins), scalable to multi-layer logic; validated in mammalian cells and safe-harbor landing pads (see below). [Nature](https://www.nature.com/articles/s41467-022-33287-9?utm_source=chatgpt.com)
* **Related:** A general blueprint for **multi-input CRISPRa promoters** (deep/wide circuits) was formalized, showing systematic design for **AND-like** and multi-branch networks (while that study optimized prokaryotic/cell-free parts, the architecture is directly portable). [PNAS](https://www.pnas.org/doi/10.1073/pnas.2220358120?utm_source=chatgpt.com)
**3) Synthetic TF toolkits (ZF-based) for orthogonal, druggable control**
* **What:** Libraries like **COMET** (44 activators, 12 repressors, **83 cognate promoters**) and **synZiFTRs** (compact, human-derived, drug-controllable zinc-finger TFs). They give you many **orthogonal inputs** you can map to sensors. [Nature](https://www.nature.com/articles/s41467-019-14147-5?utm_source=chatgpt.com), [Science](https://www.science.org/doi/10.1126/science.ade0156?utm_source=chatgpt.com)
* **Why good:** Mature, well-characterized parts; easy to compose **Boolean logic** and titratable outputs; synZiFTRs are designed with **clinical practicality** in mind (compact, humanized, small-molecule control). [Science](https://www.science.org/doi/10.1126/science.ade0156?utm_source=chatgpt.com)
**4) RNA-only, highly multiplexable logic with endoRNase switches (PERSIST)**
* **What:** **PERSIST** uses **CRISPR endoRNases** (e.g., Csy4, CasE/Cas6) as RNA-level ON/OFF switches; nine orthogonal enzymes support all 16 two-input Boolean functions; circuits resist epigenetic silencing and are good for **mRNA delivery**. [PMC](https://pmc.ncbi.nlm.nih.gov/articles/PMC9095627/?utm_source=chatgpt.com), [PubMed](https://pubmed.ncbi.nlm.nih.gov/35562172/?utm_source=chatgpt.com)
* **Why good:** Great for **layering** many inputs (miRNAs, endogenous RNAs, inducible RNAs) before a CRISPRa/TF stage. [PMC](https://pmc.ncbi.nlm.nih.gov/articles/PMC9095627/?utm_source=chatgpt.com)
**5) Extracellular marker + intracellular state AND-gates (synNotch + CRISPR/TF)**
* **What:** **synNotch** receptors convert a **surface antigen** cue into a user-defined transcriptional output (e.g., gRNA, TF, recombinase). Combine with miRNA logic to get **(antigen) AND (miRNA profile)** targeting. [Oxford Academic](https://academic.oup.com/nar/article/45/13/e118/3813634?utm_source=chatgpt.com)
* **Why good:** Adds an **environmental/cell-cell interaction** input; widely used in engineered T cells and generalizable. [Oxford Academic](https://academic.oup.com/nar/article/45/13/e118/3813634?utm_source=chatgpt.com)
**6) Memory & intersectional targeting with recombinases**
* **What:** Use **Cre/Flp/Dre/Bxb1/φC31** to “write” the result of a sensor computation (e.g., flip a STOP, install a payload). Modern sets like **BLADE** provide clean, composable recombinase logic gates. [Courses at UW](https://courses.cs.washington.edu/courses/cse599x/10sp/RNAi_circuits.pdf?utm_source=chatgpt.com)
* **Why good:** **Stable memory** of transient cues; works beautifully with enhancers/promoters and safe-harbor **landing pads** (Bxb1, etc.). [PMC](https://pmc.ncbi.nlm.nih.gov/articles/PMC10661055/?utm_source=chatgpt.com)
**7) Split effectors as AND-gates**
* **What:** Express **split-Cas9/base editors/dCas9** halves under different sensors; reconstitution (inteins, rapamycin dimerizers, light, etc.) gives a powerful **AND gate** with low leak. [Oxford Academic](https://academic.oup.com/nar/article/43/13/6450/2414291?utm_source=chatgpt.com), [PMC](https://pmc.ncbi.nlm.nih.gov/articles/PMC4503468/?utm_source=chatgpt.com), [Nature](https://www.nature.com/articles/s41467-023-41331-5?utm_source=chatgpt.com)
---
# “Starter blueprints” you can adapt
**A) All-genetic AND/NOT for cell identity (portable across systems)**
* Inputs: 2–5 **miRNAs** (identity markers) + 1 **drug** (clinician control).
* Logic layer: **miRNA-repressed Acr** (NOT), **miRNA-gated dCas9** halves (AND).
* Core: **CRISPRa responsive promoters** (2–6 orthogonal promoters) driving payload(s).
* Output: Editor/effector under a final **gRNA-programmable promoter**. [PMC](https://pmc.ncbi.nlm.nih.gov/articles/PMC6648350/?utm_source=chatgpt.com), [Nature](https://www.nature.com/articles/s41467-022-33287-9?utm_source=chatgpt.com)
**B) Antigen × state intersection**
* synNotch (antigen) → expresses **gRNA A**; **miR logic** licenses **dCas9-VPR**; final CRISPRa promoter needs **gRNA A & B** (k-of-n logic). Add **PERSIST** RNAs to expand inputs without new proteins. [Oxford Academic](https://academic.oup.com/nar/article/45/13/e118/3813634?utm_source=chatgpt.com), [PMC](https://pmc.ncbi.nlm.nih.gov/articles/PMC9095627/?utm_source=chatgpt.com)
**C) “Sense → Decide → Write” with memory**
* Sensors (miRNA, synNotch, PERSIST) → recombinase (**Bxb1/Cre**) flips in/out a **payload cassette** at a **landing pad**; optional **CRISPRoff/on** locks the state epigenetically. [PMC](https://pmc.ncbi.nlm.nih.gov/articles/PMC10661055/?utm_source=chatgpt.com), [ScienceDirect](https://www.sciencedirect.com/science/article/pii/S0092867421003536?utm_source=chatgpt.com)
---
# Where enhancers/promoters come from (and how to target cell types)
* **Enhancer discovery & delivery:** **PESCA** (parallel enhancer single-cell assay), **AAV-STARR-seq**, **scMPRA** and new **enhancer-AAV toolboxes** give vetted, neuron and brain-region selective enhancers that you can pair with the logic above. [bioRxiv](https://www.biorxiv.org/content/10.1101/570895v1.full.pdf?utm_source=chatgpt.com), [ResearchGate](https://www.researchgate.net/publication/370262377_An_unbiased_AAV-STARR-seq_screen_revealing_the_enhancer_activity_map_of_genomic_regions_in_the_mouse_brain_in_vivo?utm_source=chatgpt.com), [Cell](https://www.cell.com/neuron/abstract/S0896-6273%2825%2900349-6?utm_source=chatgpt.com), [ScienceDirect](https://www.sciencedirect.com/science/article/pii/S2589004220300729?utm_source=chatgpt.com)
* **Synthetic promoter libraries:** Designed, **short** synthetic promoters (<250 bp) tuned to stimuli provide compact, modular readouts that scale in mammalian cells. [PMC](https://pmc.ncbi.nlm.nih.gov/articles/PMC11604768/?utm_source=chatgpt.com)
---
# Emerging or reprogrammable techniques
* **Bridge RNA–guided recombinases (IS110)** — **RNA-programmable recombination** (insertions, inversions, excisions) with a **single protein + RNA**, potentially a compact way to “write” enhancer/promoter states downstream of sensors. [Nature](https://www.nature.com/articles/s41586-024-07552-4?utm_source=chatgpt.com), [PMC](https://pmc.ncbi.nlm.nih.gov/articles/PMC10849738/?utm_source=chatgpt.com)
* **RNA-sensing guide RNAs** — gRNAs that **activate only upon detecting specific RNAs** (e.g., dual-toehold or miRNA-sensing sgRNAs) offer direct **RNA→CRISPR** wiring to expand input channels without new proteins. [eLife](https://elifesciences.org/articles/87722?utm_source=chatgpt.com), [Nature](https://www.nature.com/articles/s42003-024-06988-8?utm_source=chatgpt.com)
* **Programmable, humanized synTFs** — **synZiFTRs** for small molecule in-human reprogramming; [Science](https://www.science.org/doi/10.1126/science.ade0156?utm_source=chatgpt.com)
---
# Stable and predictable integration
Use **landing pads** (e.g., **Bxb1** at AAVS1/ROSA26) to drop in circuits once, then swap modules via **RMCE**. This tames position effects, supports large libraries, and keeps expression stable over time — ideal for iterating sensor combinations. [PMC](https://pmc.ncbi.nlm.nih.gov/articles/PMC10661055/?utm_source=chatgpt.com), [ScienceDirect](https://www.sciencedirect.com/science/article/pii/S2667237522001825?utm_source=chatgpt.com)
---
# Reprogrammable cell-type targeting & logic
**miRNA sensors & CRISPR control**
* Cell-type-specific CRISPR with **miR-Cas9 switches**. [Oxford Academic](https://academic.oup.com/nar/article/45/13/e118/3813634?utm_source=chatgpt.com)
* **Cas9-ON** via **miRNA-repressed anti-CRISPR** (in vivo, multi-ortholog). [PMC](https://pmc.ncbi.nlm.nih.gov/articles/PMC6648350/?utm_source=chatgpt.com)
**CRISPRa/CRISPRi-responsive promoters in mammalian cells**
* **CRISPR-based synthetic transcription platform** with orthogonal operator/promoter libraries (predictable tuning, stable chromosomal expression). [Nature](https://www.nature.com/articles/s41467-022-33287-9?utm_source=chatgpt.com)
**Synthetic TF toolkits (large orthogonal palettes)**
* **COMET** (44 activators, 12 repressors, 83 promoters; Boolean logic). [Nature](https://www.nature.com/articles/s41467-019-14147-5?utm_source=chatgpt.com)
* **synZiFTRs** (compact, human-derived, drug-regulated). [Science](https://www.science.org/doi/10.1126/science.ade0156?utm_source=chatgpt.com)
**RNA-only logic (composability & anti-silencing)**
* **PERSIST** — 9 orthogonal endoRNases; full 2-input Boolean set; long-term stability. [PMC](https://pmc.ncbi.nlm.nih.gov/articles/PMC9095627/?utm_source=chatgpt.com)
**Extracellular antigen sensing**
* **synNotch** for combinatorial antigen recognition; wire to CRISPR/TF layers. [Oxford Academic](https://academic.oup.com/nar/article/45/13/e118/3813634?utm_source=chatgpt.com)
**Enhancer discovery & cell-type AAVs**
* **PESCA** (parallel enhancer single-cell assay). [bioRxiv](https://www.biorxiv.org/content/10.1101/570895v1.full.pdf?utm_source=chatgpt.com)
* **Enhancer-AAV toolboxes** for cortical/neuronal subtypes. [Cell](https://www.cell.com/neuron/abstract/S0896-6273%2825%2900349-6?utm_source=chatgpt.com), [bioRxiv](https://www.biorxiv.org/content/10.1101/2024.06.10.597244v3.full.pdf?utm_source=chatgpt.com)
* **AAV-STARR-seq in brain** (thousands of enhancer candidates). [ResearchGate](https://www.researchgate.net/publication/370262377_An_unbiased_AAV-STARR-seq_screen_revealing_the_enhancer_activity_map_of_genomic_regions_in_the_mouse_brain_in_vivo?utm_source=chatgpt.com)
* **scMPRA** for cell-type-specific cis-activity. [bioRxiv](https://www.biorxiv.org/content/10.1101/2021.11.11.468308v2?utm_source=chatgpt.com)
**Split-Cas and inducible editors (compact AND-gates)**
* **Split dCas9** logic (3-input AND; Suntag integration). [PMC](https://pmc.ncbi.nlm.nih.gov/articles/PMC5063958/?utm_source=chatgpt.com)
* **Rapamycin/light-inducible split-Cas9/base editors**. [PMC](https://pmc.ncbi.nlm.nih.gov/articles/PMC4503468/?utm_source=chatgpt.com), [Science](https://www.science.org/doi/10.1126/sciadv.abb1777?utm_source=chatgpt.com), [Nature](https://www.nature.com/articles/s41467-023-41331-5?utm_source=chatgpt.com)
**Landing pads & RMCE**
* **Bxb1/φC31** landing pads for reproducible expression and library scale-up. [PMC](https://pmc.ncbi.nlm.nih.gov/articles/PMC10661055/?utm_source=chatgpt.com), [ScienceDirect](https://www.sciencedirect.com/science/article/pii/S2667237522001825?utm_source=chatgpt.com)
**NEW mechanisms (compact & programmable)**
* **Bridge RNA–guided recombination (IS110)** — RNA-programmed genome rearrangements. [Nature](https://www.nature.com/articles/s41586-024-07552-4?utm_source=chatgpt.com)
---
# Practical build tips
* Start with identity sensors like miRNA MREs to gate Acr (for OFF where you don’t want editing) and/or **dCas9 halves** (for AND where you do). This is for multi-input logic in vivo. [PMC](https://pmc.ncbi.nlm.nih.gov/articles/PMC6648350/?utm_source=chatgpt.com)
* **Do logic with gRNAs, not proteins.** Use **CRISPRa/CRISPRi-responsive promoters/operators** to combine many inputs; adding a new input is just a new gRNA/operator. [Nature](https://www.nature.com/articles/s41467-022-33287-9?utm_source=chatgpt.com)
* **Keep circuits portable**: Prefer **RNA-level gates (PERSIST)** and **ZF TFs** (COMET/synZiFTRs) when payload budget or anti-silencing matters; both scale well. [PMC](https://pmc.ncbi.nlm.nih.gov/articles/PMC9095627/?utm_source=chatgpt.com), [Nature](https://www.nature.com/articles/s41467-019-14147-5?utm_source=chatgpt.com)
* **Stabilize with landing pads** (Bxb1), then **swap modules** by RMCE for rapid rewiring. [PMC](https://pmc.ncbi.nlm.nih.gov/articles/PMC10661055/?utm_source=chatgpt.com)
* **Find strong, specific enhancers** for hard cell types (brain): use **PESCA/AAV-STARR-seq/scMPRA** hits and layer logic on top (miRNA/synNotch) for extra precision. [bioRxiv](https://www.biorxiv.org/content/10.1101/570895v1.full.pdf?utm_source=chatgpt.com), [ResearchGate](https://www.researchgate.net/publication/370262377_An_unbiased_AAV-STARR-seq_screen_revealing_the_enhancer_activity_map_of_genomic_regions_in_the_mouse_brain_in_vivo?utm_source=chatgpt.com)
---
For highly reprogrammable, multi-sensor, cell-specific expression, the most buildable stack today is:
miRNA & extracellular sensors (synNotch) → RNA/TF logic (PERSIST, COMET/synZiFTR, CRISPRa/i-responsive promoters) → optional memory (recombinases, CRISPRoff) → payload, delivered via AAV/mRNA and stabilized with **Bxb1 landing pads**. This stack is modular, scalable (add inputs as gRNAs), and supported by well-used papers and toolkits. [PMC](https://pmc.ncbi.nlm.nih.gov/articles/PMC6648350/?utm_source=chatgpt.com), [Nature](https://www.nature.com/articles/s41467-022-33287-9?utm_source=chatgpt.com)
---
# Other cell-specific expression strategies
Here are several approaches to create scalable, multi-sensor cell-specific expression systems using modern gene editing technologies:
* combinatorial recombinase logic gate systems (Cre/lox, Flp/FRT, STOP cassettes)
* CRISPR-based transcriptional logic
* CellREADR
* RNA logic circuits using toehold switches and riboregulators
* cell-type specific enhancers
* just go read existing promoter/enhancer libraries for cell-type fingerprinting, like neurons, excitatory neuron, inhibitory neuron, etc...
# other stuff
allele specific sensors? "genotype sensors".
polygenic allele sensors? IQ sensor?
what about racial genotypes-- with 3 or 4 specific alleles you should be able to identify a specific racial genotype maybe?
|
