assistant: Welcome to CRISPR-GPT. I can help with the following tasks. Please select 
one to continue.

1. Meta-Mode (Step-by-Step Guidance on Pre-defined Meta-Task)
2. Auto-Mode (Customized Guidance on Free-style User Request)

Note: Chat with ChatGPT augmented with our up-to-date knowledge base 
directly by entering Questions followed by "Q:". For example, "Q: What is 
CRISPR?"


user: 1

assistant: {'Thoughts': 'The user has selected the Meta-Mode, which involves step-by-step guidance on a pre-defined meta-task. This indicates that the user prefers a structured approach with clear instructions on a specific CRISPR-related task.', 'Choice': '1'}

assistant: Please select the general gene editing scenarios to continue.
1. Generating a Knockout Using CRISPR.
2. CRISPR Base Editing Without Double-Strand Breaks.
3. Generating Small Insertion/deletion/base editing through Prime Editing.
4. Activation or Repression of Target Genes Using CRISPR.


user: 1

assistant: {'Thoughts': 'The user has selected to generate a knockout using CRISPR, which involves creating a double-strand break at a specific location in the genome to disrupt the function of a target gene.', 'Choice': '1'}

assistant: Now, let's start designing and planning your knockout experiment using CRISPR system, we will go through a step by step process as listed below:
1. Selecting a CRISPR/Cas system.
2. Selecting a delivery approach for CRISPR/Cas system.
3. Designing guideRNA for gene knockout.
4. Collecting appropriate experimantal protocols and validating the editing outcomes.


assistant: 
Step 1. Selecting a CRISPR/Cas system 

We support four types of CRISPR/Cas system for CRISPR knockout, SpCas9, enCas12a, SaCas9, and SpRYCas9, representing distinct variants with unique properties and applications.

1. SpCas9  (Jinek et al.,Science,2012;Cong et al.,Science,2013;Mali et al.,Science,2013;Jinek et al.,eLife,2013)
- Pros: Highly effective, well-studied, proven applications in many cell types/scenarios.  
- Cons: Large, potential off-target effects.  
- Uses: Genetic research, gene therapy, genetically modified models/organisms (GMMs/GMOs).

2. enCas12a  (Kleinstiver et al.,Nat Biotechnol.,2019;DeWeirdt et al.,Nat Biotechnol.,2020)
- Pros: Smaller, precise, low off-target rate, multi-target editing capability using a single guideRNA array.  
- Cons: May be less efficient than SpCas9 when using a single guideRNA (mitigated if using multi-guide array), limited target sites vs. SpCas9 due to more restrictive PAM.  
- Uses: High-precision and multiplexed genome editing (for genetic interaction, target redundant genes/pathways).

3. SaCas9  (Ran et al.,Nature,2015)
- Pros: Compact, suitable for AAV vectors (knockout genes in primary cells or in vivo).  
- Cons: limited target sites vs. SpCas9 due to more restrictive PAM.  
- Uses: In vivo gene therapy and some primary cells when using AAV delivery.

4. SpRYCas9  (Walton et al.,Science,2020)
- Pros: Wide PAM range, targets more genomic regions vs. typical SpCas9.  
- Cons: Higher off-target risks.  
- Uses: editing a specific target site where canonical PAMs for SpCas9/SaCas9/enCas12a are not available, and where specificity/off-target concerns are less critical.

*Please cite the corresponding papers if you decide to use the system.


assistant: 
Please select the CRISPR-Cas system you would like to use. Or please briefly describe your need (exp: I want to edit mouse liver/I want low off-target rate/I want to do multiplexed editing) and we could recommend the CRSIPR-Cas system to you.


user: I want to edit the mouse liver with low off-target rate

assistant: {'Thoughts': "The user wants to edit the mouse liver with a low off-target rate. SpCas9 is highly efficient but has potential for off-target effects. SaCas9 is more compact but has a narrower PAM compatibility. enCas12a offers higher precision and a lower off-target rate, which aligns with the user's requirement for low off-target effects, making it a suitable choice for precise genome editing. SpRYCas9 has broadened PAM recognition but also increased potential for off-target effects, which is not suitable given the user's need for low off-target rate.", 'Answer': 'enCas12a'} Final Result enCas12a

assistant: 
Step 2. Selecting a delivery approach for CRISPR/Cas system

Different biological systems are amenable to different delivery methods for genome-editing tools. We currently support the following delivery method:

a. Plasmid transfection
b. Lentivirus/Retrovirus delivery
c. RNP/mRNA Electroporation
d. RNP/mRNA Microinjection
e. mRNA LNP
f. AAV (Adeno-Associated Virus)


Please select one from above or please briefly describe your application cases below (exp. I hope to edit A549 cell lines/ I hope to edit mouse liver).

*Please note we currently only support mammalian cells/tissues.


user: I want to edit the mouse liver with low off-target rate

assistant: Please find the recommended delivery approach below:

Top choice: 
### f. AAV (Adeno-Associated Virus)

**Introduction**: AAV vectors are small, non-pathogenic viruses used to deliver therapeutic genes. They do not integrate into the host genome, reducing the risk of insertional mutagenesis.

- **Pros**:
  - **Low Immunogenicity**: AAVs typically induce a mild immune response, making them suitable for in vivo applications.
  - **Stable Expression**: Can provide long-term gene expression in non-dividing cells.
  - **Broad Range of Serotypes**: Different AAV serotypes can target specific tissues, enhancing tissue-specific delivery.

- **Cons**:
  - **Limited Payload Capacity**: Can carry only small genes or gene fragments, limiting its versatility.
  - **Variable Transduction Efficiency**: Efficiency can vary widely among different tissues or cell types.
  - **Cost and Production**: High costs and complex production processes can be limiting factors.

- **Applications**: Primarily used in gene therapy for treating genetic disorders, with several AAV-based therapies already in clinical use.


The relevant papers we found on Google Scholar: 
- AAV-mediated gene delivery to the mouse liver: https://link.springer.com/protocol/10.1007/978-1-4939-9065-8_12
- AAV-mediated liver-directed gene therapy: https://link.springer.com/protocol/10.1007/978-1-61779-370-7_6
- Inflammation promotes the loss of adeno-associated virus–mediated transgene expression in mouse liver: https://www.sciencedirect.com/science/article/pii/S0016508511004732
- Comparative study of liver gene transfer with AAV vectors based on natural and engineered AAV capsids: https://www.cell.com/molecular-therapy-family/molecular-therapy/fulltext/S1525-0016(16)30204-0
- Liver injury increases the incidence of HCC following AAV gene therapy in mice: https://www.cell.com/molecular-therapy-family/molecular-therapy/fulltext/S1525-0016(20)30552-9
- Inverse zonation of hepatocyte transduction with AAV vectors between mice and non-human primates: https://www.sciencedirect.com/science/article/pii/S1096719211001879
- Adeno-associated virus (AAV) capsid genes isolated from rat and mouse liver genomic DNA define two new AAV species distantly related to AAV-5: https://www.sciencedirect.com/science/article/pii/S0042682206003540
- A dual AAV system enables the Cas9-mediated correction of a metabolic liver disease in newborn mice: https://www.nature.com/articles/nbt.3469
- Systematic evaluation of AAV vectors for liver directed gene transfer in murine models: https://www.cell.com/molecular-therapy-family/molecular-therapy/fulltext/S1525-0016(16)31588-X
- AAV gene therapy corrects OTC deficiency and prevents liver fibrosis in aged OTC-knock out heterozygous mice: https://www.sciencedirect.com/science/article/pii/S1096719217300240

Second choice: 
### e. mRNA LNP

**Introduction**: Lipid nanoparticles (LNPs) are used as delivery vehicles for mRNA, protecting it from degradation and enhancing cellular uptake and endosomal escape.

- **Pros**:
  - **High Delivery Efficiency**: Offers efficient delivery of mRNA to a variety of cell types, including hard-to-transfect cells.
  - **Reduced Immunogenicity**: The use of LNPs can mitigate the innate immune response compared to naked mRNA.
  - **Versatile Application**: Suitable for vaccine development, therapeutic protein production, and gene editing.

- **Cons**:
  - **Formulation Complexity**: The development of effective LNP formulations requires significant expertise and optimization.
  - **Potential for Off-Target Effects**: While rare, there's a possibility of unintended interactions with non-target cells.
  - **Storage and Stability Issues**: mRNA-LNP formulations may require stringent storage conditions to maintain stability.

- **Applications**: COVID-19 vaccines, cancer immunotherapies, genetic diseases, and protein replacement therapies.


The relevant papers we found on Google Scholar: 
- Noninvasive imaging of lipid nanoparticle–mediated systemic delivery of small-interfering RNA to the liver: https://www.cell.com/molecular-therapy-family/molecular-therapy/fulltext/S1525-0016(16)30842-5
- Engineered ionizable lipid nanoparticles for targeted delivery of RNA therapeutics into different types of cells in the liver: https://www.science.org/doi/abs/10.1126/sciadv.abf4398
- VEGFA mRNA-LNP promotes biliary epithelial cell-to-hepatocyte conversion in acute and chronic liver diseases and reverses steatosis and fibrosis: https://www.cell.com/cell-stem-cell/fulltext/S1934-5909(23)00392-2
- Specific hepatic delivery of procollagen α1 (I) small interfering RNA in lipid‐like nanoparticles resolves liver fibrosis: https://aasldpubs.onlinelibrary.wiley.com/doi/pdf/10.1002/hep.27936
- Transient yet robust expression of proteins in the mouse liver via intravenous injection of lipid nanoparticle-encapsulated nucleoside-modified mRNA: https://bio-protocol.org/pdf/bio-protocol4184.pdf
- Engineered ionizable lipid nanoparticles mediated efficient siRNA delivery to macrophages for anti-inflammatory treatment of acute liver injury: https://www.sciencedirect.com/science/article/pii/S0378517322010444
- Cell subtypes within the liver microenvironment differentially interact with lipid nanoparticles: https://link.springer.com/article/10.1007/s12195-019-00573-4
- Lipid nanoparticle-targeted mRNA therapy as a treatment for the inherited metabolic liver disorder arginase deficiency: https://www.pnas.org/doi/abs/10.1073/pnas.1906182116
- Targeting ABCB4 using mRNA-LNP for the treatment of rare liver diseases: https://www.biorxiv.org/content/10.1101/2023.04.11.535868.abstract
- Downregulation of hepatic lipopolysaccharide binding protein improves lipogenesis-induced liver lipid accumulation: https://www.cell.com/molecular-therapy-family/nucleic-acids/fulltext/S2162-2531(22)00199-8



assistant: Which delivery method would you like to use?


user: I would like to use AAV

assistant: {'Thoughts': 'The user has expressed a preference for using Adeno-Associated Virus (AAV) as the delivery method. Considering the pros and cons of AAV vectors, they are a suitable choice for delivering therapeutic genes due to their low immunogenicity, stable expression in non-dividing cells, and the ability to target specific tissues using different serotypes. However, it is important to note the limitations in payload capacity and potential variability in transduction efficiency, which may affect the delivery of larger genes or result in inconsistent results across different tissues. Given these considerations, AAV is a viable and commonly used method for gene therapy applications.', 'Choice': 'f'}

assistant: 
Step 3. Designing guideRNA for gene knockout

Next, we would like to guide you through the design of guideRNA for your target gene. Designing an effective and specific single-guide RNA (sgRNA) is crucial for successful CRISPR/Cas9-mediated genome editing.There are several factors to consider in this process, including the selection of target sites that are unique to your gene of interest and minimizing potential off-target effects. 



assistant: 
We could help you design the guideRNA sequence based on CRISPick (https://portals.broadinstitute.org/gppx/crispick/public) (Kim et al.,Nat Biotechnol.,2018; DeWeirdt et al.,Nat Biotechnol.,2021; Doench et al.,Nat Biotechnol.,2016; Sanson et al.,Nat Commun, 2018).

Could you please describe what sgRNAs do you need? (For example, you can type in: I hope to design 2 sgRNAs targeting human EGFR; Please give me 1 sgRNA target Exon 1 of mouse Tgfbr1.)

*Please include the target and species information.
*Please cite the corresponding papers if you decide to use the designed sgRNA.


user: Could you help me design 4 sgRNAs targeting mouse Hfe?

assistant: Sugessted exon/exons: {'Exon/exons': 'Exon 2 or Exon 3', 'Reason': {'Since the user did not specify which exon to target, we can suggest targeting exons that are most likely to produce a functional knockout. Exons 2 and 3 are good candidates because they are early in the gene and mutations here can disrupt the open reading frame leading to a non-functional protein.'}}

assistant: 
Based on the information you provided, it may be beneficial to design sgRNAs targeting specific exons within the genes.

Would you like to continue sgRNA design with the exon/exons we suggested?


user: no

assistant: https://crispr-gpt.s3.amazonaws.com/fa796689-e1e0-43bf-a9dd-a7c3ec11962b?AWSAccessKeyId=AKIA6ODU3LTXQWKWTRED&Signature=37bF5%2B%2BUhHtl4lFYYzBMdloQKm0%3D&Expires=1725809559

| Target Gene Symbol   | CRISPR Mechanism   | sgRNA Sequence          | PAM Sequence   |   Exon Number |   sgRNA Cut Position (1-based) |   On-Target Efficacy Score |   Off-Target Rank |   Combined Rank |
|:---------------------|:-------------------|:------------------------|:---------------|--------------:|-------------------------------:|---------------------------:|------------------:|----------------:|
| Hfe                  | CRISPRko           | CTCCCGCACTCACGCGGTGGCAG | TTTC           |             1 |                       23710635 |                     0.8714 |                 4 |               1 |
| Hfe                  | CRISPRko           | GTGAAAGTGACTCGCCACTGGGC | TTTG           |             4 |                       23706105 |                     0.8005 |                35 |               2 |
| Hfe                  | CRISPRko           | TCCCGCACTCACGCGGTGGCAGT | TTCC           |             1 |                       23710636 |                     0.7575 |                27 |               3 |
| Hfe                  | CRISPRko           | TGGACCATCATGGGCAACTATAA | CTTC           |             2 |                       23708041 |                     0.7732 |                41 |               4 |

assistant: 
Would you like to use CRISPRitz to predict sgRNA off-target effects?


user: yes

assistant: To identify potential off-target sites for your genome editing guides, we recommend you using CRISPRitz (Cancellieri et al.,Bioinformatics,2020), a rapid, high-throughput and variant-aware in silico off-target site identification package.

Please find the github page below:
https://github.com/pinellolab/CRISPRitz

*Please cite the corresponding papers if you decide to use the tool.

Our CRISPR-GPT can run CRISPRitz on our server for you, if you'd like CRISPR-GPT to help you run this step please type “Yes”, otherwise, type “No” if you prefer to run CRISPRitz locally, and we will give you a set of instructions. Would you like CRISPR-GPT to run off-target analysis for you? (Yes/No)


user: yes

assistant: Are you hoping to perform (1). Mismatches-Only search or (2). Mismatches + Bulges search?

user: 1

assistant: 
Please provide the species (human or mouse)
1. human
2. mouse


user: 2

assistant: 
Please provide the CAS system (SpCas9 or AsCas12a)
1. SpCas9
2. AsCas12a


user: 2

assistant: 
Please provide the sequence entering guide+PAM for SpCas9 (e.g. TCACCCAGGCTGGAATACAGNNN) or PAM+guide for AsCas12a (e.g. NNNNTCACCCAGGCTGGAATACAG).


user: NNNNCTCCCGCACTCACGCGGTGGCAG

assistant: Running CRISPRItz mismatch_only with species: mouse, CAS: AsCas12a, sequence: NNNNCTCCCGCACTCACGCGGTGGCAG. We will send you an email once the result files are ready to be downloaded.

assistant: Step 4. Collecting appropriate experimantal protocols and validating the editing outcomes.

Based on information from the previous step 2, we are showing you related protocols using AAV delivery


assistant: Here are several reference papers we found on Google Scholar: 
- AAV gene therapy prevents and reverses heart failure in a murine knockout model of Barth syndrome: https://www.ahajournals.org/doi/abs/10.1161/circresaha.119.315956
- Rapid and highly efficient inducible cardiac gene knockout in adult mice using AAV-mediated expression of Cre recombinase: https://academic.oup.com/cardiovascres/article-abstract/104/1/15/317207
- Highly efficient neuronal gene knockout in vivo by CRISPR-Cas9 via neonatal intracerebroventricular injection of AAV in mice: https://www.nature.com/articles/s41434-021-00224-2
- AAV-mediated delivery of ciliary neurotrophic factor prolongs photoreceptor survival in the rhodopsin knockout mouse: https://www.cell.com/molecular-therapy-family/molecular-therapy/fulltext/S1525-0016(00)90252-1
- AAV-mediated gene therapy in the guanylate cyclase (RetGC1/RetGC2) double knockout mouse model of Leber congenital amaurosis: https://www.liebertpub.com/doi/abs/10.1089/hum.2012.193

assistant: In this protocol, you can validate the knockout through the following ways, including: 

1. Staining of target protein by antibody, and measure level of knockout using flow cytometry or western blot.
2. Mutation sequencing by Sanger Sequencing
3. Mutation sequencing by Next-generation Sequencing (NGS)

Which one would you like to use?


user: 3

assistant: [SP4.docx](https://crispr-gpt.s3.amazonaws.com/SP4.docx?AWSAccessKeyId=AKIA6ODU3LTXQWKWTRED&Signature=ZIQcQCEuopoTlThYeFsvbRrtSwo%3D&Expires=1725810388)

[BP8_NGS.docx](https://crispr-gpt.s3.amazonaws.com/BP8_NGS.docx?AWSAccessKeyId=AKIA6ODU3LTXQWKWTRED&Signature=SvlDHuRpd%2BtsFYTWlBaLetBygCo%3D&Expires=1725810388)

 
 
 (add amplicon for NGS and add Crispresso website)

assistant: We could help you design the related PCR primers using Primer3 (Untergasser et al.,Nucleic Acids Res.,2012) through following options:

1. Primer3 - command-line script to design primers - we run the script for you.
2. Primer-blast (Web tool, Manual design one by one)
3. Primer3 - command-line script to design primers - you run the code yourself

*Please cite the corresponding papers if you decide to use the tool.

Which one do you prefer?



user: 2

assistant: {'ChatResponse': {'Thoughts': 'The user has chosen option 2, which involves using Primer-blast, a web tool for manual primer design.', 'Choice': '2'}}

assistant: We suggest using the webtool: https://www.ncbi.nlm.nih.gov/tools/primer-blast/ with options:
We recommend using NCBI-PrimerBlast to design your PCR primers.
Parameter should be:
- (1) PCR Template:
Sequence: Please include 200bp upstream and 200bp downstream of the gRNA.
Range: Forward: 0-200 Reverse: 220-400
- (2) Primer Parameters:
PCR product size: 100-200bp
- (3) Primer Pair Specificity Checking Parameters:
Database: Refseq mRNA
Organism: Your model organisms.

Please don’t forget to add corresponding illumina adaptors for the downstream library construction.


assistant: 
Next, we could help you analyze the NGS data using
CRISPResso2 (Clement et al., Nat Biotechnol.,2019).

1. CRISPResso2 - command-line script to analyze data - we run the script for you.
2. CRISPResso2 (Web tool)

*Please cite the corresponding papers if you decide to use the tool.
Which one do you prefer?


user: 2

assistant: 
Please find the webtool below: http://crispresso2.pinellolab.org/submission
*Please cite the corresponding papers if you decide to use the tool.


assistant: Finished. Clear the current chat or start a new one.