CRISPRedict: a CRISPR-Cas9 web tool for interpretable efficiency predictions

Konstantakos, Vasileios; Nentidis, Anastasios; Krithara, Anastasia; Παλιούρας, Γεώργιος

doi:10.1093/nar/gkac466

Cited by 7 publications

(8 citation statements)

References 21 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Konstantakos et al . [ 102 ] introduced a new interpretable gRNA efficiency prediction model and the related web tool, called CRISPRedict, including various regression and classification models for gRNA scoring. This web tool offers accurate efficiency predictions under different experimental conditions (e.g.…”

Section: Traditional Machine Learning Models and Their Applications I...mentioning

confidence: 99%

Using traditional machine learning and deep learning methods for on- and off-target prediction in CRISPR/Cas9: a review

Sherkatghanad

Abdar

Charlier

2023

Briefings in Bioinformatics

View full text Add to dashboard Cite

CRISPR/Cas9 (Clustered Regularly Interspaced Short Palindromic Repeats and CRISPR-associated protein 9) is a popular and effective two-component technology used for targeted genetic manipulation. It is currently the most versatile and accurate method of gene and genome editing, which benefits from a large variety of practical applications. For example, in biomedicine, it has been used in research related to cancer, virus infections, pathogen detection, and genetic diseases. Current CRISPR/Cas9 research is based on data-driven models for on- and off-target prediction as a cleavage may occur at non-target sequence locations. Nowadays, conventional machine learning and deep learning methods are applied on a regular basis to accurately predict on-target knockout efficacy and off-target profile of given single-guide RNAs (sgRNAs). In this paper, we present an overview and a comparative analysis of traditional machine learning and deep learning models used in CRISPR/Cas9. We highlight the key research challenges and directions associated with target activity prediction. We discuss recent advances in the sgRNA–DNA sequence encoding used in state-of-the-art on- and off-target prediction models. Furthermore, we present the most popular deep learning neural network architectures used in CRISPR/Cas9 prediction models. Finally, we summarize the existing challenges and discuss possible future investigations in the field of on- and off-target prediction. Our paper provides valuable support for academic and industrial researchers interested in the application of machine learning methods in the field of CRISPR/Cas9 genome editing.

show abstract

Section: Traditional Machine Learning Models and Their Applications I...mentioning

confidence: 99%

Using traditional machine learning and deep learning methods for on- and off-target prediction in CRISPR/Cas9: a review

Sherkatghanad

Abdar

Charlier

2023

Briefings in Bioinformatics

View full text Add to dashboard Cite

show abstract

“…( 33 ) and Moreno-Mateos et al. ( 34 ), DeepSpCas9 ( 35 ), DeepHF (U6 promoter and T7 promoter) ( 36 ), Azimuth 2.0 ( 31 ), CRISPRedict (U6 promoter and T7 promoter) ( 37 ), Vienna Bioactivity CRISPR score (VBC) ( 38 ), CRISPick ( 39 ) and SPROUT ( 40 ). The majority of the gRNAs were designed regardless of the score calculated by any specific prediction software (Figure 1B – O ).…”

Section: Resultsmentioning

confidence: 99%

Epigenetic profiles guide improved CRISPR/Cas9-mediated gene knockout in human T cells

Ito,

Inoue,

Nakashima

et al. 2023

Nucleic Acids Research

View full text Add to dashboard Cite

Genetic modification of specific genes is emerging as a useful tool to enhance the functions of antitumor T cells in adoptive immunotherapy. Current advances in CRISPR/Cas9 technology enable gene knockout during in vitro preparation of infused T-cell products through transient transfection of a Cas9-guide RNA (gRNA) ribonucleoprotein complex. However, selecting optimal gRNAs remains a major challenge for efficient gene ablation. Although multiple in silico tools to predict the targeting efficiency have been developed, their performance has not been validated in cultured human T cells. Here, we explored a strategy to select optimal gRNAs using our pooled data on CRISPR/Cas9-mediated gene knockout in human T cells. The currently available prediction tools alone were insufficient to accurately predict the indel percentage in T cells. We used data on the epigenetic profiles of cultured T cells obtained from transposase-accessible chromatin with high-throughput sequencing (ATAC-seq). Combining the epigenetic information with sequence-based prediction tools significantly improved the gene-editing efficiency. We further demonstrate that epigenetically closed regions can be targeted by designing two gRNAs in adjacent regions. Finally, we demonstrate that the gene-editing efficiency of unstimulated T cells can be enhanced through pretreatment with IL-7. These findings enable more efficient gene editing in human T cells.

show abstract

“…We compared BoostMEC with 10 other competing models, including CRISPRon (CNN) [ 22 ], CRISPRedict (2 separate linear models, each optimized for U6 or T7 promoters) [ 8 ], DeepSpCas9 (CNN) [ 21 ], Azimuth (boosted regression trees) [ 14 , 15 ], and others [ 9 , 10 , 12 , 13 , 31 , 34 ] utilized in Haeussler et al [ 25 ]. Predictions for CRISPRon, CRISPRedict, and DeepSpCas9 were obtained by utilizing the software made available by the authors of each study.…”

Section: Resultsmentioning

confidence: 99%

“…In contrast, methods based on conventional statistical or machine learning methods, often shown to have less competitive performance, are continuing to be explored for their interpretable properties. For example, in one recent study, Konstantakos and coauthors [ 8 ] developed a prediction tool based on binomial and linear regression, CRISPRedict, which achieves competitive performance compared to other recent tools, but with the added benefit of model explainability and interpretable predictions. In this paper, we contribute a novel tool termed BoostMEC for CRISPR efficiency prediction.…”

Section: Introductionmentioning

confidence: 99%

“…Features are typically generated through one-hot encoding of sgRNA target region nucleotides and dinucleotides, obtaining k-mer counts for those regions, and through the generation of features based on thermodynamic and epigenetic properties of the sgRNAs and target regions. Model architecture choices are varied and have included linear regression [8,9], binomial regression [8], SVMs [10][11][12], elastic net [13], boosted regression trees [14,15], Bayesian ridge regression [16], multi-step models [17][18][19], convolutional neural networks (CNN) [20][21][22][23], and recurrent neural networks (RNN) [24].…”

mentioning

confidence: 99%

See 1 more Smart Citation

BoostMEC: predicting CRISPR-Cas9 cleavage efficiency through boosting models

et al. 2022

View full text Add to dashboard Cite

Background In the CRISPR-Cas9 system, the efficiency of genetic modifications has been found to vary depending on the single guide RNA (sgRNA) used. A variety of sgRNA properties have been found to be predictive of CRISPR cleavage efficiency, including the position-specific sequence composition of sgRNAs, global sgRNA sequence properties, and thermodynamic features. While prevalent existing deep learning-based approaches provide competitive prediction accuracy, a more interpretable model is desirable to help understand how different features may contribute to CRISPR-Cas9 cleavage efficiency. Results We propose a gradient boosting approach, utilizing LightGBM to develop an integrated tool, BoostMEC (Boosting Model for Efficient CRISPR), for the prediction of wild-type CRISPR-Cas9 editing efficiency. We benchmark BoostMEC against 10 popular models on 13 external datasets and show its competitive performance. Conclusions BoostMEC can provide state-of-the-art predictions of CRISPR-Cas9 cleavage efficiency for sgRNA design and selection. Relying on direct and derived sequence features of sgRNA sequences and based on conventional machine learning, BoostMEC maintains an advantage over other state-of-the-art CRISPR efficiency prediction models that are based on deep learning through its ability to produce more interpretable feature insights and predictions.

show abstract

CRISPRedict: a CRISPR-Cas9 web tool for interpretable efficiency predictions

Cited by 7 publications

References 21 publications

Using traditional machine learning and deep learning methods for on- and off-target prediction in CRISPR/Cas9: a review

Using traditional machine learning and deep learning methods for on- and off-target prediction in CRISPR/Cas9: a review

Epigenetic profiles guide improved CRISPR/Cas9-mediated gene knockout in human T cells

BoostMEC: predicting CRISPR-Cas9 cleavage efficiency through boosting models

Contact Info

Product

Resources

About