co-accessibility

Name: co-accessibility
Author: GPTomics

1.1kstars185forks

How to install co-accessibility?

gptomics/bioskills/co-accessibility

$npx -y skills add GPTomics/bioSkills --skill co-accessibility

Installs into the current project.

Summary

Infer cis-regulatory connections (peak-to-peak co-accessibility) from scATAC-seq using Cicero, ArchR getCoAccessibility, or SCENIC+. Use when linking enhancer accessibility to promoter accessibility, identifying enhancer-gene pairs from chromatin alone (without paired RNA), runni

Files · 1

View on GitHub

SKILL.md

1	## Version Compatibility
2
3	Reference examples tested with: Cicero 1.20+, monocle3 1.3+, ArchR 1.0.2+, SCENIC+ 1.0+, pycisTopic 1.0+, Signac 1.13+, GenomicRanges 1.54+, GenomicInteractions 1.36+, BSgenome.Hsapiens.UCSC.hg38 1.4+.
4
5	Verify before use:
6	- R: `packageVersion('<pkg>')` then `?function_name` to verify parameters
7	- Python: `pip show <package>` then `help(module.function)` to check signatures
8
9	If code throws unexpected errors, introspect the installed package and adapt rather than retrying.
10
11	# Co-accessibility (cis-Regulatory Linkage)
12
13	"Which enhancers connect to which promoters in my scATAC data?" -> Use cell-to-cell variability in joint accessibility of nearby peaks to infer cis-regulatory connections without explicit RNA expression. Output is a peak-pair graph with co-accessibility scores; thresholding produces enhancer-gene candidate pairs.
14
15	- R: `cicero::run_cicero(input_cds, genomic_coords)` -> peak-pair connection scores
16	- R: `ArchR::addCoAccessibility(proj)` -> ArchR-internal Cicero wrapper
17	- Python: `pycisTopic` + `SCENIC+` for network-level inference combining ATAC + RNA + motifs
18
19	Co-accessibility is NOT 3D contact; it's a statistical association based on cell-to-cell co-variation. Strong co-accessibility correlates with Hi-C/Micro-C contacts (~30-50% concordance) but is not equivalent.
20
21	## What Co-accessibility Captures vs What It Doesn't
22
23	\| Captures \| Misses \|
24	\|----------\|--------\|
25	\| Peak pairs that vary together across cell states \| 3D physical contacts that don't vary in accessibility \|
26	\| Cis-regulatory grammar within a cell type \| Trans-chromosomal interactions \|
27	\| Active enhancer-promoter pairs \| Constitutive structural contacts \|
28	\| Lineage-specific regulation \| Developmental contacts that opened before scATAC sample \|
29	\| Distance-decay biology of enhancer-promoter \| Hub enhancers that contact many distal targets \|
30
31	For physical contact, use Hi-C, Micro-C, or PCHi-C. Co-accessibility is the chromatin-only proxy.
32
33	## Algorithmic Taxonomy
34
35	\| Tool \| Method \| Input \| Output \| Strength \| Fails when \|
36	\|------\|--------\|-------\|--------\|----------\|------------\|
37	\| Cicero (Pliner 2018) \| Graphical lasso on aggregated cell metacells \| scATAC peak-cell matrix + cell trajectory \| Peak-pair connection score (0-1) \| Original, well-validated; integrates with Monocle3 \| Slow on >50K cells; sensitive to alpha tuning \|
38	\| ArchR getCoAccessibility \| Cicero-based; uses ArchR's metacell aggregation \| ArchR project \| Same as Cicero \| Built-in to ArchR pipeline; faster on large datasets \| Tied to ArchR; same biology as Cicero \|
39	\| SCENIC+ (Bravo 2023) \| Multi-step: co-accessibility + motif scoring + RNA correlation \| Multiome (ATAC + RNA) or paired \| TF-driven enhancer-gene networks \| Most comprehensive; multi-modal \| Multiome data required; computationally heavy \|
40	\| LinkPeaks (Signac) \| Pearson correlation of accessibility with paired gene expression \| Multiome \| Peak-gene linkage score \| Direct enhancer-gene from RNA correlation \| Multiome-only; not pure ATAC \|
41	\| GeneHancer / FANTOM5 / EpiMap \| Bulk-derived enhancer-gene reference \| None (database lookup) \| Pre-computed enhancer-gene pairs \| Comprehensive; published references \| Cell-type-agnostic; may not match the biology of interest \|
42
43	Methodology evolves; verify against Pliner 2018 (Cicero), Bravo 2023 (SCENIC+), Nasser 2021 (ABC model alternative for enhancer-gene), and current Hi-C concordance benchmarks.
44
45	## How Cicero Works (Conceptually)
46
47	Cell-to-cell variability is too sparse for direct correlation. Cicero solves this via metacells:
48
49	1. Reduce dimensionality (UMAP from input).
50	2. Build k-NN graph of cells.
51	3. Aggregate k cells into metacells (default k = 50).
52	4. Compute correlation in accessibility across metacells, restricted to peak pairs within `genomic_distance_max` (default 500 kb cis).
53	5. Apply graphical lasso with regularization `alpha` to sparsify the correlation matrix.
54	6. Output: per-pair connection score; positive = co-variation, negative = anti-co-variation.
55
56	Connection thresholds typically 0.05-0.5; > 0.25 is high-confidence.
57
58	## Per-Tool Failure Modes
59
60	### Cicero -- alpha tuning shifts results
61
62	Trigger: Default alpha (sometimes computed automatically from data); custom alpha < 0.5 or > 5.
63
64	Mechanism: Alpha controls graphical lasso regularization. Too low: dense graph with many spurious connections; too high: sparse with biology missing.
65
66	Symptom: Connection count varies 10-100x across alpha sweeps.
67
68	Fix: Use Cicero's `estimat

People who liked this also viewed

6 picks

Type

co-accessibility

1.1kstars185forks

View on GitHub

How to install co-accessibility?

gptomics/bioskills/co-accessibility

$npx -y skills add GPTomics/bioSkills --skill co-accessibility

Installs into the current project.

Summary

Files · 1

View on GitHub

SKILL.md

1	## Version Compatibility
2
3	Reference examples tested with: Cicero 1.20+, monocle3 1.3+, ArchR 1.0.2+, SCENIC+ 1.0+, pycisTopic 1.0+, Signac 1.13+, GenomicRanges 1.54+, GenomicInteractions 1.36+, BSgenome.Hsapiens.UCSC.hg38 1.4+.
4
5	Verify before use:
6	- R: `packageVersion('<pkg>')` then `?function_name` to verify parameters
7	- Python: `pip show <package>` then `help(module.function)` to check signatures
8
9	If code throws unexpected errors, introspect the installed package and adapt rather than retrying.
10
11	# Co-accessibility (cis-Regulatory Linkage)
12
13	"Which enhancers connect to which promoters in my scATAC data?" -> Use cell-to-cell variability in joint accessibility of nearby peaks to infer cis-regulatory connections without explicit RNA expression. Output is a peak-pair graph with co-accessibility scores; thresholding produces enhancer-gene candidate pairs.
14
15	- R: `cicero::run_cicero(input_cds, genomic_coords)` -> peak-pair connection scores
16	- R: `ArchR::addCoAccessibility(proj)` -> ArchR-internal Cicero wrapper
17	- Python: `pycisTopic` + `SCENIC+` for network-level inference combining ATAC + RNA + motifs
18
19	Co-accessibility is NOT 3D contact; it's a statistical association based on cell-to-cell co-variation. Strong co-accessibility correlates with Hi-C/Micro-C contacts (~30-50% concordance) but is not equivalent.
20
21	## What Co-accessibility Captures vs What It Doesn't
22
23	\| Captures \| Misses \|
24	\|----------\|--------\|
25	\| Peak pairs that vary together across cell states \| 3D physical contacts that don't vary in accessibility \|
26	\| Cis-regulatory grammar within a cell type \| Trans-chromosomal interactions \|
27	\| Active enhancer-promoter pairs \| Constitutive structural contacts \|
28	\| Lineage-specific regulation \| Developmental contacts that opened before scATAC sample \|
29	\| Distance-decay biology of enhancer-promoter \| Hub enhancers that contact many distal targets \|
30
31	For physical contact, use Hi-C, Micro-C, or PCHi-C. Co-accessibility is the chromatin-only proxy.
32
33	## Algorithmic Taxonomy
34
35	\| Tool \| Method \| Input \| Output \| Strength \| Fails when \|
36	\|------\|--------\|-------\|--------\|----------\|------------\|
37	\| Cicero (Pliner 2018) \| Graphical lasso on aggregated cell metacells \| scATAC peak-cell matrix + cell trajectory \| Peak-pair connection score (0-1) \| Original, well-validated; integrates with Monocle3 \| Slow on >50K cells; sensitive to alpha tuning \|
38	\| ArchR getCoAccessibility \| Cicero-based; uses ArchR's metacell aggregation \| ArchR project \| Same as Cicero \| Built-in to ArchR pipeline; faster on large datasets \| Tied to ArchR; same biology as Cicero \|
39	\| SCENIC+ (Bravo 2023) \| Multi-step: co-accessibility + motif scoring + RNA correlation \| Multiome (ATAC + RNA) or paired \| TF-driven enhancer-gene networks \| Most comprehensive; multi-modal \| Multiome data required; computationally heavy \|
40	\| LinkPeaks (Signac) \| Pearson correlation of accessibility with paired gene expression \| Multiome \| Peak-gene linkage score \| Direct enhancer-gene from RNA correlation \| Multiome-only; not pure ATAC \|
41	\| GeneHancer / FANTOM5 / EpiMap \| Bulk-derived enhancer-gene reference \| None (database lookup) \| Pre-computed enhancer-gene pairs \| Comprehensive; published references \| Cell-type-agnostic; may not match the biology of interest \|
42
43	Methodology evolves; verify against Pliner 2018 (Cicero), Bravo 2023 (SCENIC+), Nasser 2021 (ABC model alternative for enhancer-gene), and current Hi-C concordance benchmarks.
44
45	## How Cicero Works (Conceptually)
46
47	Cell-to-cell variability is too sparse for direct correlation. Cicero solves this via metacells:
48
49	1. Reduce dimensionality (UMAP from input).
50	2. Build k-NN graph of cells.
51	3. Aggregate k cells into metacells (default k = 50).
52	4. Compute correlation in accessibility across metacells, restricted to peak pairs within `genomic_distance_max` (default 500 kb cis).
53	5. Apply graphical lasso with regularization `alpha` to sparsify the correlation matrix.
54	6. Output: per-pair connection score; positive = co-variation, negative = anti-co-variation.
55
56	Connection thresholds typically 0.05-0.5; > 0.25 is high-confidence.
57
58	## Per-Tool Failure Modes
59
60	### Cicero -- alpha tuning shifts results
61
62	Trigger: Default alpha (sometimes computed automatically from data); custom alpha < 0.5 or > 5.
63
64	Mechanism: Alpha controls graphical lasso regularization. Too low: dense graph with many spurious connections; too high: sparse with biology missing.
65
66	Symptom: Connection count varies 10-100x across alpha sweeps.
67
68	Fix: Use Cicero's `estimat

People who liked this also viewed

6 picks

Type