SNIPSNP

SNIPSNP is an advanced computational web tool (part of the crisprtools.org suite) designed to automate the creation of Homology-Directed Repair (HDR) templates. Whether you are correcting existing genetic mutations or introducing new ones, SNIPSNP streamlines the design process by intelligently introducing synonymous variants into your template—effectively preventing Cas nucleases from re-cutting the DNA after a successful edit.

Key features of the tool include:

• Intelligent Optimization: Choose between Safety First, Disruption First, or Balanced strategies to dictate how the algorithm introduces protective synonymous variants.
• Complex Edits Made Simple: Input multiple variants simultaneously within a 50-bp window and easily toggle whether they should be repaired, retained, or newly introduced.
• Advanced Predictive Modeling: Evaluate guide efficiency and safety using the CRISPR-MFH algorithm, and predict cellular variant risks using alphaGenome Splicing Predictions.
• Comprehensive Analytics: Visualize your edits on an interactive genomic map, assess the toxicity of added variants using CADD and dbSNP data, and automatically generate lab-ready primers and probes to verify your edits.

(Note: SNIPSNP seamlessly integrates with the CHOPOFF tool, allowing you to instantly perform deep-dive off-target analyses on your chosen guides).

SNIPSNP

A Complete Guide to Using SNIPSNP for HDR Template Design

SNIPSNP is a powerful computational tool available on crisprtools.org designed to help researchers build Homology-Directed Repair (HDR) templates. Whether you want to repair existing genetic variants or create entirely new ones, SNIPSNP automates the process of designing templates and adding synonymous variants to prevent Cas nucleases from re-cutting the edited DNA.

This guide covers all the features, inputs, and analytical tools available within the SNIPSNP platform.

1. Getting Started: Access and Setup

To access all features, start by clicking Login with ORCID. Logging in allows you to interact with the website (such as voting on features) and, most importantly, saves your search history. Once approved, you will see notifications and your past jobs.

When starting a new design, always assign a clear Job Name (e.g., "Example Job One"). Because complex jobs can take up to 30 minutes to run, giving them an identifiable name ensures you can easily find them in your history and view the results later without having to run them again. If you ever need a quick refresher on the UI, you can click the Instructions button or the Tutorial button.

2. Core Configuration

Before inputting variants, you need to establish the basic parameters of your design.

Optimization Schemes

This is the most important selection. It dictates how the algorithm chooses extra synonymous variants to add to your template. These extra variants disrupt the guide RNA's binding site to prevent re-cutting. You have three options: * Balanced (Default): The recommended choice. It provides an optimal balance between patient safety, PAM disruption, and variant quality. * Safety First: Ideal if you are editing real human cells. This strategy strictly avoids non-coding overlaps, prefers known-benign variants, and prioritizes genome safety over maximal guide disruption. * Disruption First: Focuses heavily on disrupting the guide first, with safety as an afterthought. This is useful for model organisms where absolute patient safety is not the primary concern.

Species and Chromosome Selection

Select your target Species. While SNIPSNP supports multiple species, the Human and Mouse pipelines are fully featured. These organisms have enhanced design capabilities because the tool can access comprehensive, specific annotation databases like dbSNP.

Next, specify the Chromosome (you can type to search or select from a dropdown, which also includes extra loci). All variants you input must reside on this selected chromosome.

3. Inputting Variants (The Core Feature)

One of the most powerful features of SNIPSNP is the ability to input multiple variants at once, provided they fall within a short 50-base-pair window. For example, if you sequenced a patient and found a cluster of variants in a specific locus, you can input every single one of them. Variants must be inputted using the official VCF/dbSNP formatting perspective (from the sense / plus strand).

For each variant, you input the Reference allele (automatically fetched) and the Alternate allele (what is replacing it, e.g., "T" replacing "C").

Crucially, you must use the toggle switches to tell the algorithm where this variant exists: * Genome ON / Template ON: The variant currently exists in the patient's genome (so guides must be designed assuming the mutation is there). You also want this variant to remain present on the template after the HDR edit. * Genome ON / Template OFF: The variant exists in the patient's genome, but you want to fix it. The template will carry the clean reference sequence to replace the variant. * Genome OFF / Template ON: The genome is currently clean/reference, but you want to introduce this variant via the HDR template (e.g., creating a designer organism).

Handling Indels (Insertions and Deletions): If you are inputting an insertion (e.g., Reference "G" becomes Alternate "GAGA") or a large chunk deletion, the variant must be anchored to a specific reference base. In the insertion example, "G" at position 120 is the anchor base, and "AGA" is the actual insert.

4. Advanced Design Options

SNIPSNP offers deep customization for your template architecture and predictive models:

• Distance to Cut & Max Variants: By default, the tool designs a maximum of 3 extra synonymous variants. It is best to use the least amount of extra variants possible to do the job; adding too many mutates the genome in unexpected ways.
• Template & Homology Arm Size: The default HDR template size is 120 base pairs. By default, 30 base pairs on each end are protected as Homology Recombination arms. These protected outer regions will not be edited with extra variants.
• Guide Filtering: You can restrict the design engine to build templates for one specific, concrete guide.
• Intron Exclusion: By default, the tool excludes 6 base pairs at intron-exon boundaries to prevent the accidental disruption of splicing mechanics.
• Primers and Probes: Disabled by default because they are computationally time-consuming to design. (However, they are pre-enabled when loading the example dataset).
• CRISPR-MFH Algorithm: This is our proprietary predictive model for guide off-target interactions and guide mismatch disruption. You can disable it if you prefer older, handcrafted rule based system.
• alphaGenome Splicing Predictions: Predicts variant risks based on cellular context. Warning: alphaGenome is not licensed for commercial applications. Non-commercial users can optimize predictions by selecting their specific cellular context (e.g., T-cell, HEK cell, bone, etc.). To disable it entirely, leave the field empty.

5. Interpreting the Results

Normal jobs take time, but if a job has been run previously (or if you click "Load Example"), the pre-calculated results load instantly.

The Genomic Context View

At the top of the results is the Genomic Context View. This visual map is highly interactive and connected to the data tables below. If you hover over or click a template in the table (e.g., Template "CCN_3"), it will highlight the corresponding guide ("CCN3") and visually isolate the specific variants designed for that template. You can zoom in and out to inspect the loci.

Evaluating Guides (Safety First)

When reviewing the data tables, it is highly recommended to look at the Guides table first. Safety is paramount, and you want a guide with high on-target efficiency and minimal off-targets. * The top 5 safest guides are highlighted by default. * You can sort tables by clicking the columns (hold Control/Command to sort by multiple columns simultaneously). * Guides are scored by efficiency predictors (Doench 2014, Moreno Mateos, Labuhn 2018) and ranked. * Off-targets are summarized by exact distance (Distance 0, Distance 1, Distance 2) and whether they overlap known genes.

Deep Dive into Guides (CHOPOFF Tool): By clicking "View Details" on any guide, you enter the CHOPOFF tool interface. Here, you can see if a guide has known lab activity. It provides exact CRISPR-MFH alignment scores, showing the precise alignments that create the off-target distance. It also lists the specific genes overlapped by off-targets, providing direct links to the NCBI Gene Card, Uniprot, Ensemble, and the UCSC Genome Browser. You can even click "Design Primers" for specific off-target sites to generate 5 lab-ready primers to test that specific off-target risk experimentally.

The Editing Strategy: The "Zero-Variant" Template

Sometimes, the variant you are intentionally editing is enough to disable the guide on its own.

For example, look at Guide "CCN_1". It is a highly safe guide with very few off-targets. It happens to physically overlap the variant we want to fix. Because our HDR template will remove the patient's mutation and restore the reference sequence, the guide loses its binding match. Therefore, SNIPSNP will recommend a template for CCN_1 with zero extra synonymous variants designed for it, because your primary edit inherently disables the guide.

Evaluating Extra Variants

If your chosen guide does require extra synonymous variants, you must evaluate what you are introducing into the cell using the Variant Scores tables: * CADD Score: Predicts toxicity (a score of 0 means low toxicity). * dbSNP Priority Tier: Tier 1 means the exact variant and allele already exists naturally and safely in the human population. Tier 2 means a variant exists at that specific spot in the population, but with a different allele (e.g., a "Y" instead of an "A"). * Frameshifts: If your primary intended edit creates a frameshift, the entire gene is essentially disabled. In this scenario, downstream metrics like codon number or synonymous status will simply read as "NA" (Not Available), because the gene is already broken.

Exporting and Verification (Primers/Probes)

At the bottom of the results, you will find the Primers (marked in blue) designed for the whole loci to help you confirm that your overall editing was successful. You will also find Probes designed specifically for each Guide + Template combination to detect if the specific edit worked in the lab.

Every data table includes an Info (i) button explaining each column's metrics, as well as a Download button so you can export the data as a CSV for your records.

With this information, you can copy your optimized template sequences and move confidently into the lab to start your experiments!

This tutorial was created with the grant from COST Action: CA21113, Reference: E-COST-GRANT-CA21113-55536e26.