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

Precise measurement of CRISPR genome editing outcomes through single-cell DNA sequencing

Gene therapy for clinical applications necessitates a comprehensive, accurate, and precise measurement of gene-edited drug products. State-of-the-art pipelines for evaluating editing outcomes rely primarily on bulk sequencing approaches, which are limited to population-level assessment. Here, we leveraged Tapestri, a single-cell sequencing technology for an in-depth analysis of editing outcomes. Using this platform, we characterized the genotype of triple-edited cells simultaneously at more than 100 loci, including editing zygosity, structural variations, and cell clonality. Our findings revealed a unique editing pattern in nearly every edited cell, highlighting the importance of single-cell resolution measurement to ensure the highest safety standards.

A comprehensive review of neurotransmitter modulation via artificial intelligence: A new frontier in personalized neurobiochemistry

The deployment of artificial intelligence (AI) is revolutionizing neuropharmacology and drug development, allowing the modulation of neurotransmitter systems at the personal level. This review focuses on the neuropharmacology and regulation of neurotransmitters using predictive modeling, closed-loop neuromodulation, and precision drug design. The fusion of AI with applications such as machine learning, deep-learning, and even computational modeling allows for the real-time tracking and enhancement of biological processes within the body. An exemplary application of AI is the use of DeepMind's AlphaFold to design new GABA reuptake inhibitors for epilepsy and anxiety. Likewise, Benevolent AI and IBM Watson have fast-tracked drug repositioning for neurodegenerative conditions. Furthermore, we identified new serotonin reuptake inhibitors for depression through AI screening. In addition, the application of Deep Brain Stimulation (DBS) settings using AI for patients with Parkinson's disease and for patients with major depressive disorder (MDD) using reinforcement learning-based transcranial magnetic stimulation (TMS) leads to better treatment. This review highlights other challenges including algorithm bias, ethical concerns, and limited clinical validation. Their proposal to incorporate AI with optogenetics, CRISPR, neuroprosthesis, and other advanced technologies fosters further exploration and refinement of precision neurotherapeutic approaches. By bridging computational neuroscience with clinical applications, AI has the potential to revolutionize neuropharmacology and improve patient-specific treatment strategies. We addressed critical challenges, such as algorithmic bias and ethical concerns, by proposing bias auditing, diverse datasets, explainable AI, and regulatory frameworks as practical solutions to ensure equitable and transparent AI applications in neurotransmitter modulation.

Application of CRISPR-Cas9 in microbial cell factories

Metabolically engineered bacterial strains are rapidly emerging as a pivotal focus in the biosynthesis of an array of bio-based ingredients. Presently, CRISPR (clustered regularly interspaced short palindromic repeats) and its associated RNA-guided endonuclease (Cas9) are regarded as the most promising tool, having ushered in a transformative advancement in genome editing. Because of CRISPR-Cas9's accuracy and adaptability, metabolic engineers are now able to create novel regulatory systems, optimize pathways more effectively, and make extensive genome-scale alterations. Nevertheless, there are still obstacles to overcome in the application of CRISPR-Cas9 in novel microorganisms, particularly those industrial microorganism hosts that are resistant to traditional genetic manipulation techniques. How to further extend CRISPR-Cas9 to these microorganisms is an urgent problem to be solved. This article first introduces the mechanism and application of CRISPR-Cas9, and then discusses how to optimize CRISPR-Cas9 as a genome editing tool, including how to reduce off-target effects and how to improve targeting efficiency by optimizing design. Through offering a comprehensive perspective on the revolutionary effects of CRISPR-Cas9 in microbial cell factories, we hope to stimulate additional research and development in this exciting area.

An Overview and Comparative Analysis of CRISPR-SpCas9 gRNA Activity Prediction Tools

Design of guide RNA (gRNA) with high efficiency and specificity is vital for successful application of the CRISPR gene editing technology. Although many machine learning (ML) and deep learning (DL)-based tools have been developed to predict gRNA activities, a systematic and unbiased evaluation of their predictive performance is still needed. Here, we provide a brief overview of in silico tools for CRISPR design and assess the CRISPR datasets and statistical metrics used for evaluating model performance. We benchmark seven ML and DL-based CRISPR-Cas9 editing efficiency prediction tools across nine CRISPR datasets covering six cell types and three species. The DL models CRISPRon and DeepHF outperform the other models exhibiting greater accuracy and higher Spearman correlation coefficient across multiple datasets. We compile all CRISPR datasets and in silico prediction tools into a GuideNet resource web portal, aiming to facilitate and streamline the sharing of CRISPR datasets. Furthermore, we summarize features affecting CRISPR gene editing activity, providing important insights into model performance and the further development of more accurate CRISPR prediction models.

Artificial Intelligence and Computational Biology in Gene Therapy: A Review

One of the trending fields in almost all areas of science and technology is artificial intelligence. Computational biology and artificial intelligence can help gene therapy in many steps including: gene identification, gene editing, vector design, development of new macromolecules and modeling of gene delivery. There are various tools used by computational biology and artificial intelligence in this field, such as genomics, transcriptomic and proteomics data analysis, machine learning algorithms and molecular interaction studies. These tools can introduce new gene targets, novel vectors, optimized experiment conditions, predict the outcomes and suggest the best solutions to avoid undesired immune responses following gene therapy treatment.

CRISPR-MFH: A Lightweight Hybrid Deep Learning Framework with Multi-Feature Encoding for Improved CRISPR-Cas9 Off-Target Prediction

BACKGROUND

The CRISPR-Cas9 system has emerged as one of the most promising gene-editing technologies in biology. However, off-target effects remain a significant challenge. While recent advances in deep learning have led to the development of models for off-target prediction, these models often fail to fully leverage sequence pair information. Furthermore, as the models' parameter sizes increase, so do their complexities, limiting their practical applicability.

METHODS

In this study, we introduce a novel multi-feature independent encoding method, which encodes the gRNA-DNA sequence pair into three distinct feature matrices to minimize information loss. Additionally, we propose a lightweight hybrid deep learning framework, CRISPR-MFH, that integrates multi-scale separable convolutions and hybrid attention mechanisms for efficient and accurate off-target prediction.

RESULTS

Extensive experiments across multiple benchmark datasets demonstrate that the proposed encoding method effectively captures critical features and that CRISPR-MFH outperforms or matches state-of-the-art models with significantly fewer parameters across multiple evaluation metrics.

CONCLUSIONS

This study offers a novel perspective for advancing deep learning technology in the realm of CRISPR-Cas9 off-target detection.

Engineering a New Generation of Gene Editors: Integrating Synthetic Biology and AI Innovations

CRISPR-Cas technology has revolutionized biology by enabling precise DNA and RNA edits with ease. However, significant challenges remain for translating this technology into clinical applications. Traditional protein engineering methods, such as rational design, mutagenesis screens, and directed evolution, have been used to address issues like low efficacy, specificity, and high immunogenicity. These methods are labor-intensive, time-consuming, and resource-intensive and often require detailed structural knowledge. Recently, computational strategies have emerged as powerful solutions to these limitations. Using artificial intelligence (AI) and machine learning (ML), the discovery and design of novel gene-editing enzymes can be streamlined. AI/ML models predict activity, specificity, and immunogenicity while also enhancing mutagenesis screens and directed evolution. These approaches not only accelerate rational design but also create new opportunities for developing safer and more efficient genome-editing tools, which could eventually be translated into the clinic.

Microbial production systems and optimization strategies of antimicrobial peptides: a review

Antibiotic resistance has become a public safety issue of the twenty-first century, posing a growing threat and drawing increased attention. Compared to traditional antibiotics, antimicrobial peptides (AMPs), as naturally produced small peptides, can target multiple pathways within pathogens and render them less prone to developing resistance. This makes them promising alternatives to antibiotics. However, traditional chemical synthesis methods face challenges, such as high costs, low yields, and poor stability, limiting the large-scale industrial production of AMPs. Despite extensive research to improve AMP production efficiency, issues such as low yields and complex extraction processes continue to pose significant barriers to commercial application. Therefore, there is an urgent need for new biosynthesis strategies and optimization methods to enhance AMP production efficiency and quality. This review summarizes the sources, classification, mechanisms of action and recent advances in the microbial synthesis of AMPs. It also explores innovative production methods, including recombinant microbial expression systems, fusion tags, codon optimization, tandem multimer expression, and hybrid peptide expression. Furthermore, we review the applications of gene editing technologies and artificial intelligence in AMP production, providing new perspectives and strategies for efficient, large-scale AMP production.

Transitioning from wet lab to artificial intelligence: a systematic review of AI predictors in CRISPR

The revolutionary CRISPR-Cas9 system leverages a programmable guide RNA (gRNA) and Cas9 proteins to precisely cleave problematic regions within DNA sequences. This groundbreaking technology holds immense potential for the development of targeted therapies for a wide range of diseases, including cancers, genetic disorders, and hereditary diseases. CRISPR-Cas9 based genome editing is a multi-step process such as designing a precise gRNA, selecting the appropriate Cas protein, and thoroughly evaluating both on-target and off-target activity of the Cas9-gRNA complex. To ensure the accuracy and effectiveness of CRISPR-Cas9 system, after the targeted DNA cleavage, the process requires careful analysis of the resultant outcomes such as indels and deletions. Following the success of artificial intelligence (AI) in various fields, researchers are now leveraging AI algorithms to catalyze and optimize the multi-step process of CRISPR-Cas9 system. To achieve this goal AI-driven applications are being integrated into each step, but existing AI predictors have limited performance and many steps still rely on expensive and time-consuming wet-lab experiments. The primary reason behind low performance of AI predictors is the gap between CRISPR and AI fields. Effective integration of AI into multi-step CRISPR-Cas9 system demands comprehensive knowledge of both domains. This paper bridges the knowledge gap between AI and CRISPR-Cas9 research. It offers a unique platform for AI researchers to grasp deep understanding of the biological foundations behind each step in the CRISPR-Cas9 multi-step process. Furthermore, it provides details of 80 available CRISPR-Cas9 system-related datasets that can be utilized to develop AI-driven applications. Within the landscape of AI predictors in CRISPR-Cas9 multi-step process, it provides insights of representation learning methods, machine and deep learning methods trends, and performance values of existing 50 predictive pipelines. In the context of representation learning methods and classifiers/regressors, a thorough analysis of existing predictive pipelines is utilized for recommendations to develop more robust and precise predictive pipelines.

Gene therapy for genetic diseases: challenges and future directions

Genetic diseases constitute the majority of rare human diseases, resulting from abnormalities in an individual's genetic composition. Traditional treatments offer limited relief for these challenging conditions. In contrast, the rapid advancement of gene therapy presents significant advantages by directly addressing the underlying causes of genetic diseases, thereby providing the potential for precision treatment and the possibility of curing these disorders. This review aims to delineate the mechanisms and outcomes of current gene therapy approaches in clinical applications across various genetic diseases affecting different body systems. Additionally, genetic muscular disorders will be examined as a case study to investigate innovative strategies of novel therapeutic approaches, including gene replacement, gene suppression, gene supplementation, and gene editing, along with their associated advantages and limitations at both clinical and preclinical levels. Finally, this review emphasizes the existing challenges of gene therapy, such as vector packaging limitations, immunotoxicity, therapy specificity, and the subcellular localization and immunogenicity of therapeutic cargos, while discussing potential optimization directions for future research. Achieving delivery specificity, as well as long-term effectiveness and safety, will be crucial for the future development of gene therapies targeting genetic diseases.

DeepMEns: an ensemble model for predicting sgRNA on-target activity based on multiple features

The CRISPR/Cas9 system developed from Streptococcus pyogenes (SpCas9) has high potential in gene editing. However, its successful application is hindered by the considerable variability in target efficiencies across different single guide RNAs (sgRNAs). Although several deep learning models have been created to predict sgRNA on-target activity, the intrinsic mechanisms of these models are difficult to explain, and there is still scope for improvement in prediction performance. To overcome these issues, we propose an ensemble interpretable model termed DeepMEns based on deep learning to predict sgRNA on-target activity. By using five different training and validation datasets, we constructed five sub-regressors, each comprising three parts. The first part uses one-hot encoding, wherein 0-1 representation of the secondary structure is used as the input to the convolutional neural network (CNN) with Transformer encoder. The second part uses the DNA shape feature matrix as the input to the CNN with Transformer encoder. The third part uses positional encoding feature matrices as the proposed input into a long short-term memory network with an attention mechanism. These three parts are concatenated through the flattened layer, and the final prediction result is the average of the five sub-regressors. Extensive benchmarking experiments indicated that DeepMEns achieved the highest Spearman correlation coefficient for 6 of 10 independent test datasets as compared to previous predictors, this finding confirmed that DeepMEns can accomplish state-of-the-art performance. Moreover, the ablation analysis also indicated that the ensemble strategy may improve the performance of the prediction model.

Predicting CRISPR-Cas9 off-target effects in human primary cells using bidirectional LSTM with BERT embedding

Motivation

Clustered Regularly Interspaced Short Palindromic Repeats (CRISPR)-Cas9 system is a ground-breaking genome editing tool, which has revolutionized cell and gene therapies. One of the essential components involved in this system that ensures its success is the design of an optimal single-guide RNA (sgRNA) with high on-target cleavage efficiency and low off-target effects. This is challenging as many conditions need to be considered, and empirically testing every design is time-consuming and costly. In silico prediction using machine learning models provides high-performance alternatives.

Results

We present CrisprBERT, a deep learning model incorporating a Bidirectional Encoder Representations from Transformers (BERT) architecture to provide a high-dimensional embedding for paired sgRNA and DNA sequences and Bidirectional Long Short-term Memory networks for learning, to predict the off-target effects of sgRNAs utilizing only the sgRNAs and their paired DNA sequences. We proposed doublet stack encoding to capture the local energy configuration of the Cas9 binding and applied the BERT model to learn the contextual embedding of the doublet pairs. Our results showed that the new model achieved better performance than state-of-the-art deep learning models regarding single split and leave-one-sgRNA-out cross-validations as well as independent testing.

Availability and implementation

The CrisprBERT is available at GitHub: https://github.com/OSsari/CrisprBERT.

Balanced Training Sets Improve Deep Learning-Based Prediction of CRISPR sgRNA Activity

CRISPR-Cas systems have transformed the field of synthetic biology by providing a versatile method for genome editing. The efficiency of CRISPR systems is largely dependent on the sequence of the constituent sgRNA, necessitating the development of computational methods for designing active sgRNAs. While deep learning-based models have shown promise in predicting sgRNA activity, the accuracy of prediction is primarily governed by the data set used in model training. Here, we trained a convolutional neural network (CNN) model and a large language model (LLM) on balanced and imbalanced data sets generated from CRISPR-Cas12a screening data for the yeast Yarrowia lipolytica and evaluated their ability to predict high- and low-activity sgRNAs. We further tested whether prediction performance can be improved by training on imbalanced data sets augmented with synthetic sgRNAs. Lastly, we demonstrated that adding synthetic sgRNAs to inherently imbalanced CRISPR-Cas9 data sets from Y. lipolytica and Komagataella phaffii leads to improved performance in predicting sgRNA activity, thus underscoring the importance of employing balanced training sets for accurate sgRNA activity prediction.

Learning to quantify uncertainty in off-target activity for CRISPR guide RNAs

CRISPR-based genome editing technologies have revolutionised the field of molecular biology, offering unprecedented opportunities for precise genetic manipulation. However, off-target effects remain a significant challenge, potentially leading to unintended consequences and limiting the applicability of CRISPR-based genome editing technologies in clinical settings. Current literature predominantly focuses on point predictions for off-target activity, which may not fully capture the range of possible outcomes and associated risks. Here, we present crispAI, a neural network architecture-based approach for predicting uncertainty estimates for off-target cleavage activity, providing a more comprehensive risk assessment and facilitating improved decision-making in single guide RNA (sgRNA) design. Our approach makes use of the count noise model Zero Inflated Negative Binomial (ZINB) to model the uncertainty in the off-target cleavage activity data. In addition, we present the first-of-its-kind genome-wide sgRNA efficiency score, crispAI-aggregate, enabling prioritization among sgRNAs with similar point aggregate predictions by providing richer information compared to existing aggregate scores. We show that uncertainty estimates of our approach are calibrated and its predictive performance is superior to the state-of-the-art in silico off-target cleavage activity prediction methods. The tool and the trained models are available at https://github.com/furkanozdenn/crispr-offtarget-uncertainty.

Unraveling the significance of AGPAT4 for the pathogenesis of endometriosis via a multi-omics approach

Endometriosis is characterized by the ectopic proliferation of endometrial cells, posing considerable diagnostic and therapeutic challenges. Our study investigates AGPAT4's involvement in endometriosis pathogenesis, aiming to unveil new therapeutic targets. Our investigation by analyzing eQTL data from GWAS for preliminary screening. Subsequently, within the GEO dataset, we utilized four machine learning algorithms to precisely identify risk-associated genes. Gene validity was confirmed through five Mendelian Randomization methods. AGPAT4 expression was measured by Single-Cell Analysis, ELISA and immunohistochemistry. We investigated AGPAT4's effect on endometrial stromal cells using RNA interference, assessing cell proliferation, invasion, and migration with CCK8, wound-healing, and transwell assays. Protein expression was analyzed by western blot, and AGPAT4 interactions were explored using AutoDock. Our investigation identified 11 genes associated with endometriosis risk, with AGPAT4 and COMT emerging as pivotal biomarkers through machine learning analysis. AGPAT4 exhibited significant upregulation in both ectopic tissues and serum samples from patients with endometriosis. Reduced expression of AGPAT4 was observed to detrimentally impact the proliferation, invasion, and migration capabilities of endometrial stromal cells, concomitant with diminished expression of key signaling molecules such as Wnt3a, β-Catenin, MMP-9, and SNAI2. Molecular docking analyses further underscored a substantive interaction between AGPAT4 and Wnt3a.Our study highlights AGPAT4's key role in endometriosis, influencing endometrial stromal cell behavior, and identifies AGPAT4 pathways as promising therapeutic targets for this condition.

Overcoming CRISPR-Cas9 off-target prediction hurdles: A novel approach with ESB rebalancing strategy and CRISPR-MCA model

The off-target activities within the CRISPR-Cas9 system remains a formidable barrier to its broader application and development. Recent advancements have highlighted the potential of deep learning models in predicting these off-target effects, yet they encounter significant hurdles including imbalances within datasets and the intricacies associated with encoding schemes and model architectures. To surmount these challenges, our study innovatively introduces an Efficiency and Specificity-Based (ESB) class rebalancing strategy, specifically devised for datasets featuring mismatches-only off-target instances, marking a pioneering approach in this realm. Furthermore, through a meticulous evaluation of various One-hot encoding schemes alongside numerous hybrid neural network models, we discern that encoding and models of moderate complexity ideally balance performance and efficiency. On this foundation, we advance a novel hybrid model, the CRISPR-MCA, which capitalizes on multi-feature extraction to enhance predictive accuracy. The empirical results affirm that the ESB class rebalancing strategy surpasses five conventional methods in addressing extreme dataset imbalances, demonstrating superior efficacy and broader applicability across diverse models. Notably, the CRISPR-MCA model excels in off-target effect prediction across four distinct mismatches-only datasets and significantly outperforms contemporary state-of-the-art models in datasets comprising both mismatches and indels. In summation, the CRISPR-MCA model, coupled with the ESB rebalancing strategy, offers profound insights and a robust framework for future explorations in this field.

CRISPR challenges in clinical developments

CRISPR-Cas (clustered regularly interspaced short palindromic repeats and associated proteins) is a novel genome editing technology with potential applications in treating diseases. Currently, its use in humans is restricted to clinical trials, although its growth rate is significant, and some have received initial FDA approval. It is crucial to examine and address the challenges for this technology to be implemented in clinical settings. This review aims to identify and explore new research ideas to increase of CRISPR's efficiency in treating genetic diseases and cancer, as well as its future prospects. Given that a substantial amount of previous research has focused on CRISPR-Cas delivery strategies and materials, this overview introduces specific conditions and strategies. It also discusses some of the challenges and opportunities in this field, offering a unique perspective.

New developments in biotechnology applied to microorganisms

EFSA was requested by the European Commission (in accordance with Article 29 of Regulation (EC) No 178/2002) to provide a scientific opinion on the application of new developments in biotechnology (new genomic techniques, NGTs) to viable microorganisms and products of category 4 to be released into the environment or placed on the market as or in food and feed, and to non-viable products of category 3 to be placed on the market as or in food and feed. A horizon scanning exercise identified a variety of products containing microorganisms obtained with NGTs (NGT-Ms), falling within the remit of EFSA, that are expected to be placed on the (EU) market in the next 10 years. No novel potential hazards/risks from NGT-Ms were identified as compared to those obtained by established genomic techniques (EGTs), or by conventional mutagenesis. Due to the higher efficiency, specificity and predictability of NGTs, the hazards related to the changes in the genome are likely to be less frequent in NGT-Ms than those modified by EGTs and conventional mutagenesis. It is concluded that EFSA guidances are 'partially applicable', therefore on a case-by-case basis for specific NGT-Ms, fewer requirements may be needed. Some of the EFSA guidances are 'not sufficient' and updates are recommended. Because possible hazards relate to genotypic and phenotypic changes introduced and not to the method used for the modification, it is recommended that any new guidance should take a consistent risk assessment approach for strains/products derived from or produced with microorganisms obtained with conventional mutagenesis, EGTs or NGTs.

Analysis of Emerging Variants of Turkey Reovirus using Machine Learning

Avian reoviruses continue to cause disease in turkeys with varied pathogenicity and tissue tropism. Turkey enteric reovirus has been identified as a causative agent of enteritis or inapparent infections in turkeys. The new emerging variants of turkey reovirus, tentatively named turkey arthritis reovirus (TARV) and turkey hepatitis reovirus (THRV), are linked to tenosynovitis/arthritis and hepatitis, respectively. Turkey arthritis and hepatitis reoviruses are causing significant economic losses to the turkey industry. These infections can lead to poor weight gain, uneven growth, poor feed conversion, increased morbidity and mortality and reduced marketability of commercial turkeys. To combat these issues, detecting and classifying the types of reoviruses in turkey populations is essential. This research aims to employ clustering methods, specifically K-means and Hierarchical clustering, to differentiate three types of turkey reoviruses and identify novel emerging variants. Additionally, it focuses on classifying variants of turkey reoviruses by leveraging various machine learning algorithms such as Support Vector Machines, Naive Bayes, Random Forest, Decision Tree, and deep learning algorithms, including convolutional neural networks (CNNs). The experiments use real turkey reovirus sequence data, allowing for robust analysis and evaluation of the proposed methods. The results indicate that machine learning methods achieve an average accuracy of 92%, F1-Macro of 93% and F1-Weighted of 92% scores in classifying reovirus types. In contrast, the CNN model demonstrates an average accuracy of 85%, F1-Macro of 71% and F1-Weighted of 84% scores in the same classification task. The superior performance of the machine learning classifiers provides valuable insights into reovirus evolution and mutation, aiding in detecting emerging variants of pathogenic TARVs and THRVs.

Gene editing technology to improve antitumor T-cell functions in adoptive immunotherapy

Adoptive immunotherapy, in which tumor-reactive T cells are prepared in vitro for adoptive transfer to the patient, can induce an objective clinical response in specific types of cancer. In particular, chimeric antigen receptor (CAR)-redirected T-cell therapy has shown robust responses in hematologic malignancies. However, its efficacy against most of the other tumors is still insufficient, which remains an unmet medical need. Accumulating evidence suggests that modifying specific genes can enhance antitumor T-cell properties. Epigenetic factors have been particularly implicated in the remodeling of T-cell functions, including changes to dysfunctional states such as terminal differentiation and exhaustion. Genetic ablation of key epigenetic molecules prevents the dysfunctional reprogramming of T cells and preserves their functional properties.Clustered, regularly interspaced, short palindromic repeats (CRISPR)/CRISPR-associated protein (Cas)-based gene editing is a valuable tool to enable efficient and specific gene editing in cultured T cells. A number of studies have already identified promising targets to improve the therapeutic efficacy of CAR-T cells using genome-wide or focused CRISPR screening. In this review, we will present recent representative findings on molecular insights into T-cell dysfunction and how genetic modification contributes to overcoming it. We will also discuss several technical advances to achieve efficient gene modification using the CRISPR and other novel platforms.

Interpretable CRISPR/Cas9 off-target activities with mismatches and indels prediction using BERT

Off-target effects of CRISPR/Cas9 can lead to suboptimal genome editing outcomes. Numerous deep learning-based approaches have achieved excellent performance for off-target prediction; however, few can predict the off-target activities with both mismatches and indels between single guide RNA (sgRNA) and target DNA sequence pair. In addition, data imbalance is a common pitfall for off-target prediction. Moreover, due to the complexity of genomic contexts, generating an interpretable model also remains challenged. To address these issues, firstly we developed a BERT-based model called CRISPR-BERT for enhancing the prediction of off-target activities with both mismatches and indels. Secondly, we proposed an adaptive batch-wise class balancing strategy to combat the noise exists in imbalanced off-target data. Finally, we applied a visualization approach for investigating the generalizable nucleotide position-dependent patterns of sgRNA-DNA pair for off-target activity. In our comprehensive comparison to existing methods on five mismatches-only datasets and two mismatches-and-indels datasets, CRISPR-BERT achieved the best performance in terms of AUROC and PRAUC. Besides, the visualization analysis demonstrated how implicit knowledge learned by CRISPR-BERT facilitates off-target prediction, which shows potential in model interpretability. Collectively, CRISPR-BERT provides an accurate and interpretable framework for off-target prediction, further contributes to sgRNA optimization in practical use for improved target specificity in CRISPR/Cas9 genome editing. The source code is available at https://github.com/BrokenStringx/CRISPR-BERT.

CRISPR-DIPOFF: an interpretable deep learning approach for CRISPR Cas-9 off-target prediction

CRISPR Cas-9 is a groundbreaking genome-editing tool that harnesses bacterial defense systems to alter DNA sequences accurately. This innovative technology holds vast promise in multiple domains like biotechnology, agriculture and medicine. However, such power does not come without its own peril, and one such issue is the potential for unintended modifications (Off-Target), which highlights the need for accurate prediction and mitigation strategies. Though previous studies have demonstrated improvement in Off-Target prediction capability with the application of deep learning, they often struggle with the precision-recall trade-off, limiting their effectiveness and do not provide proper interpretation of the complex decision-making process of their models. To address these limitations, we have thoroughly explored deep learning networks, particularly the recurrent neural network based models, leveraging their established success in handling sequence data. Furthermore, we have employed genetic algorithm for hyperparameter tuning to optimize these models' performance. The results from our experiments demonstrate significant performance improvement compared with the current state-of-the-art in Off-Target prediction, highlighting the efficacy of our approach. Furthermore, leveraging the power of the integrated gradient method, we make an effort to interpret our models resulting in a detailed analysis and understanding of the underlying factors that contribute to Off-Target predictions, in particular the presence of two sub-regions in the seed region of single guide RNA which extends the established biological hypothesis of Off-Target effects. To the best of our knowledge, our model can be considered as the first model combining high efficacy, interpretability and a desirable balance between precision and recall.

Challenges of CRISPR/Cas-Based Cell Therapy for Type 1 Diabetes: How Not to Engineer a "Trojan Horse"

Type 1 diabetes mellitus (T1D) is an autoimmune disease caused by the destruction of insulin-producing β-cells in the pancreas by cytotoxic T-cells. To date, there are no drugs that can prevent the development of T1D. Insulin replacement therapy is the standard care for patients with T1D. This treatment is life-saving, but is expensive, can lead to acute and long-term complications, and results in reduced overall life expectancy. This has stimulated the research and development of alternative treatments for T1D. In this review, we consider potential therapies for T1D using cellular regenerative medicine approaches with a focus on CRISPR/Cas-engineered cellular products. However, CRISPR/Cas as a genome editing tool has several drawbacks that should be considered for safe and efficient cell engineering. In addition, cellular engineering approaches themselves pose a hidden threat. The purpose of this review is to critically discuss novel strategies for the treatment of T1D using genome editing technology. A well-designed approach to β-cell derivation using CRISPR/Cas-based genome editing technology will significantly reduce the risk of incorrectly engineered cell products that could behave as a "Trojan horse".