---
title: CRISPR inputs and parameters
description: >-
  Gene targets, PAM types, scoring models, mismatch thresholds, and screen
  analysis options for all CRISPR tools.
tags:
  - crispr
  - inputs
---

You provide a target gene, guide sequence, genomic region, or screen count table; for guide design, the on-target scoring model and your GC and score filters most shape which guides rise to the top.

## CRISPR inputs and parameters

## Common input formats

Most CRISPR tools accept one of these input types:

| Input | Format | Example | Accepted By |
|-------|--------|---------|-------------|
| Gene symbol | Standard HGNC symbol | `BRCA1`, `TP53`, `KRAS` | `crispr_design_guides` |
| Guide sequence | 20-nt DNA string | `ATCGATCGATCGATCGATCG` | `crispr_offtarget` |
| Genomic region | chr:start-end | `chr17:7668402-7687550` | `crispr_design_guides` |
| Count table | CSV/TSV file | Read count matrix | `crispr_screen_analyze` |

## Guide RNA Design parameters

| Parameter | Type | Required | Default | Description |
|-----------|------|----------|---------|-------------|
| `gene` | string | Yes* | -- | Target gene symbol (HGNC) |
| `sequence` | string | Yes* | -- | Target DNA sequence (alternative to gene) |
| `pam` | string | No | `"NGG"` | PAM sequence for the Cas nuclease |
| `guide_length` | integer | No | `20` | Guide RNA length in nucleotides (17-25) |
| `organism` | string | No | `"human"` | Target organism |
| `min_gc` | number | No | `20` | Minimum GC content percentage (0-100) |
| `max_gc` | number | No | `80` | Maximum GC content percentage (0-100) |
| `min_score` | number | No | `0.0` | Minimum on-target score filter (0-1) |
| `max_guides` | integer | No | `20` | Maximum number of guides to return (1-100) |
| `exon_filter` | string | No | -- | Restrict to specific exon (e.g., `"exon3"`) |

*One of `gene` or `sequence` is required.

### Supported PAM sequences

| PAM | Nuclease | Notes |
|-----|----------|-------|
| `NGG` | SpCas9 | Default; most commonly used |
| `NNGRRT` | SaCas9 | Smaller nuclease, suitable for AAV delivery |
| `TTTV` | AsCas12a (Cpf1) | Staggered cut, AT-rich PAM |
| `NGAN` | SpCas9-VQR | Engineered variant for broader targeting |
| Custom | Any | Provide any IUPAC-encoded PAM string |

### IUPAC nucleotide codes

| Code | Bases | Code | Bases |
|------|-------|------|-------|
| `N` | A, C, G, T | `R` | A, G |
| `V` | A, C, G | `Y` | C, T |
| `H` | A, C, T | `K` | G, T |
| `D` | A, G, T | `M` | A, C |
| `B` | C, G, T | `S` | G, C |
| | | `W` | A, T |

### On-target scoring model

The guide design tool uses the Rule Set 2 algorithm (Doench et al., 2016) for on-target activity prediction. The score considers:

- Nucleotide identity at each position of the 30-mer context (guide + flanking)
- GC content of the guide
- Melting temperature
- Position-specific dinucleotide features

Scores range from 0 to 1. Guides with scores above 0.5 are generally considered active; scores above 0.7 indicate high predicted efficiency.

## Off-Target Analysis parameters

| Parameter | Type | Required | Default | Description |
|-----------|------|----------|---------|-------------|
| `guides` | string[] | Yes | -- | Guide RNA sequences to analyze (1-10, each 17-25 nt) |
| `genome` | string | No | `"GRCh38"` | Reference genome assembly |
| `max_mismatches` | integer | No | `3` | Maximum mismatches to report (0-5) |
| `pam` | string | No | `"NGG"` | PAM sequence (filters off-target sites) |
| `include_cds_only` | boolean | No | `false` | Only report off-targets in coding sequences |

### Supported reference genomes

| Genome | Organism | Assembly |
|--------|----------|----------|
| `GRCh38` | Human | GRCh38/hg38 |
| `GRCm39` | Mouse | GRCm39/mm39 |
| `Rnor_6.0` | Rat | Rnor_6.0/rn6 |

### Mismatch scoring

Off-target sites are scored based on mismatch position relative to the PAM:

| Region | Positions | Impact |
|--------|-----------|--------|
| Seed region | 1-12 (PAM-proximal) | Mismatches here strongly reduce cutting |
| Non-seed region | 13-20 (PAM-distal) | Mismatches tolerated more readily |

A guide with zero 1-mismatch off-target sites and fewer than five 2-mismatch sites in coding regions is generally considered safe for most applications.

## Screen Analysis parameters

| Parameter | Type | Required | Default | Description |
|-----------|------|----------|---------|-------------|
| `count_table` | file | Yes | -- | Read count matrix (CSV/TSV with gene, guide, sample columns) |
| `control_samples` | string[] | Yes | -- | Column names for control/reference samples |
| `treatment_samples` | string[] | Yes | -- | Column names for treatment/experimental samples |
| `direction` | string | No | `"both"` | Analysis direction: `"negative"` (depletion), `"positive"` (enrichment), `"both"` |
| `fdr_threshold` | number | No | `0.25` | FDR cutoff for significant hits (0-1) |
| `min_guides` | integer | No | `2` | Minimum guides per gene for inclusion |
| `normalization` | string | No | `"median"` | Normalization method: `"median"`, `"total"`, `"none"` |

### Count table format

The count table should be a CSV or TSV file with the following structure:

| Column | Description |
|--------|-------------|
| `gene` | Gene symbol targeted by the guide |
| `guide` | Guide RNA identifier or sequence |
| Sample columns | Raw read counts for each sample (one column per replicate) |

Example:

```
gene,guide,control_rep1,control_rep2,treatment_rep1,treatment_rep2
TP53,guide_1,1500,1400,200,180
TP53,guide_2,1200,1300,150,160
BRCA1,guide_1,800,900,750,820
```

### Analysis direction

| Direction | Identifies | Use Case |
|-----------|-----------|----------|
| `negative` | Genes whose guides are depleted in treatment | Essential genes, synthetic lethal partners |
| `positive` | Genes whose guides are enriched in treatment | Resistance genes, tumor suppressors |
| `both` | Both depleted and enriched genes | Comprehensive screen analysis |

## Read next