bio-crispr-screens-library-design

Designs pooled sgRNA libraries for CRISPR knockout, CRISPRi, CRISPRa, Cas12a, and editor screens.

Updated Sep 15, 2026
One-click install
npx skills add https://github.com/mrsonord2240/openscience-specialists --skill bio-crispr-screens-library-design-mrsonord2240
Or copy as Structured Prompt for Agent▼
Please help me install this Agent Skill.
Skill: bio-crispr-screens-library-design
Source: https://github.com/mrsonord2240/openscience-specialists/tree/main/specialists/crispr-screen-analyst/versions/1.0.0/package/skills/bio-crispr-screens-library-design
Command: npx skills add https://github.com/mrsonord2240/openscience-specialists --skill bio-crispr-screens-library-design-mrsonord2240

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill requires biopython, pandas, numpy.

What problem does it solve? Designing a pooled CRISPR screen library requires choosing the right chemistry, scoring guides for on-target activity and off-target liability, positioning them correctly relative to exons or transcription start sites, adding proper controls, and validating the cloned pool—mistakes in any step silently ruin the screen. ## Core Features & Use Cases - Chemistry and library selection: Decision tables for Cas9 KO, CRISPRi, CRISPRa, Cas12a multiplex, base-editor, and prime-editor screens, mapped to canonical libraries (Brunello, TKOv3, Dolcetto, Calabrese, Inzolia). - Guide scoring and selection: On-target scoring (Rule Set 2, DeepSpCas9, CRISPRon), off-target scoring (CFD, MIT) via CRISPOR, plus Python functions for candidate enumeration, exon-position filtering, and independent-guide selection. - TSS-relative positioning: CRISPRi (-50 to +300) and CRISPRa (-150 to -75) windows anchored to FANTOM5 CAGE peaks, with fallback handling for poorly annotated TSSs. - Controls, oligo design, and QC: Control-guide composition (~1% NTCs, safe-harbor, CEGv2/NEGv1), BsmBI cloning oligo construction, and plasmid-pool QC thresholds (Gini <0.1, skew <5, ≥99% guide detection). - Use Case: Ask the agent to design a Brunello-style Cas9 knockout library for 250 kinases with 4 guides per gene and 500 non-targeting controls, and receive a ranked guide table, oligo synthesis order, and coverage calculations. ## Quick Start Ask the agent to design a focused Cas9 knockout library for your gene list with 4 guides per gene, CFD off-target filtering, and standard non-targeting and essential-gene controls.

Frequently Asked Questions about bio-crispr-screens-library-design

High-intent search queries and answers about installing and using this skill.

FAQPage Schema
How do I design a CRISPR knockout library for a custom gene list?▼

Enumerate NGG-adjacent 20-nt protospacers in each gene's coding sequence, filter to the first 5-65% of the protein with GC 30-70% and no poly-T runs, score on-target and off-target (CFD), then greedily select 4 independent guides per gene plus ~1% non-targeting controls.

Brunello vs GeCKOv2 vs TKOv3: which genome-wide library should I use?▼

Brunello is the modern standard, built with Rule Set 2 on-target scoring and CFD off-target filtering at 4 guides per gene. TKOv3 is optimized for BAGEL2 analysis, while GeCKOv2 predates Rule Set 1 and is mainly relevant for legacy datasets.

Can I use the Azimuth Python package for Rule Set 2 scoring?▼

No. The original MicrosoftResearch/azimuth PyPI package is un-ported Python 2 and raises a SyntaxError on import under Python 3. Use Broad CRISPick, the R crisprScore::getAzimuthScores() function, or the built-in GC-composition heuristic instead.

Why does my CRISPRi screen show weak knockdown even on essential genes?▼

The most common cause is guides positioned against the wrong TSS. CRISPRi knockdown is maximal within +25 to +75 of the empirical FANTOM5 CAGE peak; canonical Ensembl/RefSeq TSSs can be off by kilobases, so re-derive TSSs from matched CAGE or GRO-seq data.

What QC thresholds should a cloned plasmid pool meet before screening?▼

Sequence the plasmid pool at 200-500 reads per sgRNA and require at least 99% guide detection, Gini coefficient below 0.1, skew ratio under 5, and fewer than 0.5% zero-count guides. Failing these thresholds indicates PCR bias, synthesis dropouts, or cloning bottlenecks.

When should I use Cas12a multiplex instead of Cas9 single-knockout?▼

Use enAsCas12a 4-guide arrays (Inzolia, in4mer) when studying paralog-buffered genes, since single Cas9 knockout misses paralog-redundant essentials. Include singleton controls for each paralog so genetic interaction can be scored as double-KO LFC minus the sum of single-KO LFCs.