Programmable Live-Cell CRISPR Imaging with Toehold-Switch-Mediated Strand Displacement

The widespread application of CRISPR-Cas9 has transformed genome engineering. Nevertheless, the precision to control the targeting activity of Cas9 requires further improvement. We report a toehold-switch-based approach to engineer the conformation of single guide RNA (sgRNA) for programmable activation of Cas9. This activation circuit is responsive to multiple inputs and can regulate the conformation of the sgRNA through toehold-switch-mediated strand displacement. We demonstrate the orthogonal suppression and activation of Cas9 with orthogonal DNA inputs. Combination of toehold switches leads to a variety of intracellular Cas9 activation programs with simultaneous and orthogonal responses, through which multiple genome loci are displayed in different colors in a controllable manner. This approach provides a new route for programing CRISPR in living cells for genome imaging and engineering.     

Application of machine learning algorithms to screen potential biomarkers under cadmium exposure based on human urine metabolic profiles

Exposure to environmental cadmium increases the health risk of residents. Early urine metabolic detection using high-resolution mass spectrometry and machine learning algorithms would be advantageous to predict the adverse health effects. Here, we conducted machine learning approaches to screen potential biomarkers under cadmium exposure in 403 urine samples. In positive and negative ionization mode, 4207 and 3558 features were extracted, respectively. We compared seven machine learning algorithms and found that the extreme gradient boosting (XGBoost) and random forest (RF) classifiers showed better accuracy and predictive performance than others. Following 5-fold cross-validation, the value of area under curve (AUC) was both 0.93 for positive and negative ionization modes in XGBoost classifier. In the RF classifier, AUC were 0.80 and 0.84 for positive and negative ionization modes, respectively. We then identified a biomarker panel based on XGBoost and RF classifiers. The incorporation of machine learning models into urine analysis using high-resolution mass spectrometry could allow a convenient assessment of cadmium exposure.     

Recent progress on discovery and properties prediction of energy materials:Simple machine learning meets complex quantum chemistry

In nature,the properties of matter are ultimately governed by the electronic structures.Quantum chemistry(QC)at electronic level matches well with a few simple physical assumptions in solving simple problems.To date,machine learning(ML)algorithm has been migrated to this field to simplify calculations and improve fidelity.This review introduces the basic information on universal electron structures of emerging energy materials and ML algorithms involved in the prediction of material properties.Then,the structure-property relationships based on ML algorithm and QC theory are reviewed.Especially,the summary of recently reported applications on classifying crystal structure,modeling electronic structure,optimizing experimental method,and predicting performance is provided.Last,an outlook on ML assisted QC calculation towards identifying emerging energy materials is also presented.     

A DNA Origami-Based Gene Editing System for Efficient Gene Therapy in Vivo

DNA nanostructures have played an important role in the development of novel drug delivery systems. Herein, we report a DNA origami-based CRISPR/Cas9 gene editing system for efficient gene therapy in vivo. In our design, a PAM-rich region precisely organized on the surface of DNA origami can easily recruit and load sgRNA/Cas9 complex by PAM-guided assembly and pre-designed DNA/RNA hybridization. After loading the sgRNA/Cas9 complex, the DNA origami can be further rolled up by the locking strands with a disulfide bond. With the incorporation of DNA aptamer and influenza hemagglutinin (HA) peptide, the cargo-loaded DNA origami can realize the targeted delivery and effective endosomal escape. After reduction by GSH, the opened DNA origami can release the sgRNA/Cas9 complex by RNase H cleavage to achieve a pronounced gene editing of a tumor-associated gene for gene therapy in vivo. This rationally developed DNA origami-based gene editing system presents a new avenue for the development of gene therapy.     

基于分子描述符和机器学习方法预测和虚拟筛选MMP-13对MMP-1的选择性抑制剂

基质金属蛋白酶-13 (MMP-13)为预防和治疗骨关节炎(OA)提供了充满希望的靶标. 通过抑制剂来阻断MMP-13的活性将会对治疗OA疾病产生潜在的作用. 然而,宽谱抑制剂同样抑制MMP家族的其它成员,特别是MMP-1,这将会导致肌与骨的综合症. 因此,设计和发现潜在的MMP-13 相对于MMP-1 的高效选择性抑制剂,在对治疗OA新型药物的研发中具有相当重要的现实意义. 本研究通过两种机器学习方法(ML)：支持向量机(SVM)和随机森林(RF)来建立分类模型,用于预测不同结构的MMP-13 对MMP-1 的选择性抑制剂. 所建这些模型的预测效果都已经达到了令人满意的精度. 在这两种ML模型中,RF对于MMP-13选择性抑制剂和非抑制剂的精度分别达到97.58%和100%. 同时,与MMP-13对MMP-1的选择性抑制最相关的分子描述符也基于不同的特征选择方法被两种模型挑选出来. 最后,用预测效果最好的RF模型虚拟筛选了ZINC数据库的“fragment-like”子集,从而得到了一系列潜在的候选药物. 研究表明,机器学习方法,特别是RF方法,对于发现潜在的MMP-13选择性抑制剂十分有效. 同时还得到了一些与MMP-13的选择性抑制相关的分子描述符.     

Computational and data driven molecular material design assisted by low scaling quantum mechanics calculations and machine learning

Electronic structure methods based on quantum mechanics (QM) are widely employed in the computational predictions of the molecular properties and optoelectronic properties of molecular materials. The computational costs of these QM methods, ranging from density functional theory (DFT) or time-dependent DFT (TDDFT) to wave-function theory (WFT), usually increase sharply with the system size, causing the curse of dimensionality and hindering the QM calculations for large sized systems such as long polymer oligomers and complex molecular aggregates. In such cases, in recent years low scaling QM methods and machine learning (ML) techniques have been adopted to reduce the computational costs and thus assist computational and data driven molecular material design. In this review, we illustrated low scaling ground-state and excited-state QM approaches and their applications to long oligomers, self-assembled supramolecular complexes, stimuli-responsive materials, mechanically interlocked molecules, and excited state processes in molecular aggregates. Variable electrostatic parameters were also introduced in the modified force fields with the polarization model. On the basis of QM computational or experimental datasets, several ML algorithms, including explainable models, deep learning, and on-line learning methods, have been employed to predict the molecular energies, forces, electronic structure properties, and optical or electrical properties of materials. It can be conceived that low scaling algorithms with periodic boundary conditions are expected to be further applicable to functional materials, perhaps in combination with machine learning to fast predict the lattice energy, crystal structures, and spectroscopic properties of periodic functional materials.

Low scaling quantum mechanics calculations and machine learning can be employed to efficiently predict the molecular energies, forces, and optical and electrical properties of molecular materials and their aggregates.     

Self‐Assembled DNA Nanoclews for the Efficient Delivery of CRISPR–Cas9 for Genome Editing

CRISPR–Cas9 represents a promising platform for genome editing, yet means for its safe and efficient delivery remain to be fully realized. A novel vehicle that simultaneously delivers the Cas9 protein and single guide RNA (sgRNA) is based on DNA nanoclews, yarn‐like DNA nanoparticles that are synthesized by rolling circle amplification. The biologically inspired vehicles were efficiently loaded with Cas9/sgRNA complexes and delivered the complexes to the nuclei of human cells, thus enabling targeted gene disruption while maintaining cell viability. Editing was most efficient when the DNA nanoclew sequence and the sgRNA guide sequence were partially complementary, offering a design rule for enhancing delivery. Overall, this strategy provides a versatile method that could be adapted for delivering other DNA‐binding proteins or functional nucleic acids.     

Non‐Viral CRISPR/Cas Gene Editing In Vitro and In Vivo Enabled by Synthetic Nanoparticle Co‐Delivery of Cas9 mRNA and sgRNA

CRISPR/Cas is a revolutionary gene editing technology with wide‐ranging utility. ¹ The safe, non‐viral delivery of CRISPR/Cas components would greatly improve future therapeutic utility. ^1e We report the synthesis and development of zwitterionic amino lipids (ZALs) that are uniquely able to (co)deliver long RNAs including Cas9 mRNA and sgRNAs. ZAL nanoparticle (ZNP) delivery of low sgRNA doses (15 nm ) reduces protein expression by >90 % in cells. In contrast to transient therapies (such as RNAi), we show that ZNP delivery of sgRNA enables permanent DNA editing with an indefinitely sustained 95 % decrease in protein expression. ZNP delivery of mRNA results in high protein expression at low doses in vitro (<600 pM) and in vivo (1 mg kg⁻¹). Intravenous co‐delivery of Cas9 mRNA and sgLoxP induced expression of floxed tdTomato in the liver, kidneys, and lungs of engineered mice. ZNPs provide a chemical guide for rational design of long RNA carriers, and represent a promising step towards improving the safety and utility of gene editing.     

On the numerical stability of two widely used PLS algorithms

Ideally, the score vectors numerically computed by an orthogonal scores partial least squares (PLS) algorithm should be orthogonal close to machine precision. However, this is not ensured without taking special precautions. The progressive loss of orthogonality with increasing number of components is illustrated for two widely used PLS algorithms, i.e., one that can be considered as a standard PLS algorithm, and SIMPLS. It is shown that the original standard PLS algorithm outperforms the original SIMPLS in terms of numerical stability. However, SIMPLS is confirmed to perform much better in terms of speed. We have investigated reorthogonalization as the special precaution to ensure orthogonality close to machine precision. Since the increase of computing time is relatively small for SIMPLS, we therefore recommend SIMPLS with reorthogonalization. Copyright © 2008 John Wiley & Sons, Ltd.     

Evolving interpretable structure-activity relationship models. 2. Using multiobjective optimization to derive multiple models

A multiobjective evolutionary algorithm (MOEA) is described for evolving multiple structure-activity relationships (SARs). The SARs are encoded in easy-to-interpret reduced graph queries which describe features that are preferentially present in active compounds compared to inactives. The MOEA addresses a limitation associated with many machine learning methods; that is, the inherent tradeoff that exists in recall and precision which is usually handled by combining the two objectives into a single measure with a consequent loss of control. By simultaneously optimizing recall and precision, the MOEA generates a family of SARs that lie on the precision-recall (PR) curve. The user is then able to select a query with an appropriate balance in the two objectives: for example, a low recall-high precision query may be preferred when establishing the SAR, whereas a high recall-low precision query may be more appropriate in a virtual screening context. Each query on the PR curve aims at capturing the structure-activity information into a single representation, and each can be considered as an alternative (equally valid) solution. We then investigate combining individual queries into teams with the aim of capturing multiple SARs that may exist in a data set, for example, as is commonly seen in high-throughput screening data sets. Team formation is carried out iteratively as a postprocessing step following the evolution of the individual queries. The inclusion of uniqueness as a third objective within the MOEA provides an effective way of ensuring the queries are complementary in the active compounds they describe. Substantial improvements in both recall and precision are seen for some data sets. Furthermore, the resulting queries provide more detailed structure-activity information than is present in a single query.     

基于机器学习方法的丙型肝炎病毒非结构蛋白5B聚合酶抑制剂活性预测

在丙型肝炎病毒(HCV)的基因复制和蛋白质成熟的过程中, 非结构蛋白5B(NS5B)作为RNA依赖的RNA聚合酶起到了重要的作用. 抑制NS5B聚合酶可以阻止丙型肝炎病毒的RNA复制, 因此成为一种治疗丙型肝炎的有效方法. 通过计算机方法进行虚拟筛选和预测NS5B聚合酶抑制剂已经变得越来越重要. 本文主要采用机器学习方法(支持向量机(SVM)、k-最近相邻法(k-NN)和C4.5决策树(C4.5 DT))对已知的丙型肝炎病毒NS5B蛋白酶抑制剂与非抑制剂建立分类预测模型. 1248个结构多样性化合物(552个NS5B抑制剂与696个非NS5B抑制剂)被用于测试分类预测系统, 并用递归变量消除法选择与NS5B抑制剂相关的性质描述符以提高预测精度. 独立验证集的总预测精度为84.1%-85.0%, NS5B抑制剂的预测精度为81.4%-91.7%, 非NS5B抑制剂的预测精度为78.2%-87.2%. 其中支持向量机给出最好的NS5B抑制剂预测精度(91.7%); C4.5决策树给出最好的非NS5B抑制剂预测精度(87.2%); k-最近相邻法给出最好的总预测精度(85.0%). 研究表明机器学习方法可以有效预测未知数据集中潜在的NS5B抑制剂, 并有助于发现与其相关的分子描述符.     

Evolution of Support Vector Machine and Regression Modeling in Chemoinformatics and Drug Discovery

Journal of Computer-Aided Molecular Design - The support vector machine (SVM) algorithm is one of the most widely used machine learning (ML) methods for predicting active compounds and molecular...     

Reaction-based machine learning representations for predicting the enantioselectivity of organocatalysts

Hundreds of catalytic methods are developed each year to meet the demand for high-purity chiral compounds. The computational design of enantioselective organocatalysts remains a significant challenge, as catalysts are typically discovered through experimental screening. Recent advances in combining quantum chemical computations and machine learning (ML) hold great potential to propel the next leap forward in asymmetric catalysis. Within the context of quantum chemical machine learning (QML, or atomistic ML), the ML representations used to encode the three-dimensional structure of molecules and evaluate their similarity cannot easily capture the subtle energy differences that govern enantioselectivity. Here, we present a general strategy for improving molecular representations within an atomistic machine learning model to predict the DFT-computed enantiomeric excess of asymmetric propargylation organocatalysts solely from the structure of catalytic cycle intermediates. Mean absolute errors as low as 0.25 kcal mol⁻¹ were achieved in predictions of the activation energy with respect to DFT computations. By virtue of its design, this strategy is generalisable to other ML models, to experimental data and to any catalytic asymmetric reaction, enabling the rapid screening of structurally diverse organocatalysts from available structural information.

A machine learning model for enantioselectivity prediction using reaction-based molecular representations.     

ML or NJ-MCL? A comparison between two robust phylogenetic methods

Large-scale gene sequencing gives an opportunity to reconstruct the tree of life and histories of multigene species phylogenies from very large datasets. A primary need for reconstructing large-scale phylogenies is a computationally efficient and accurate method. Current efforts to achieve such a goal include NJ-MCL² described by Tamura et al. (2004; 2007), an algorithm based on maximum likelihood (ML) and neighbor joining (NJ) algorithms. Although it has been reported that the NJ-MCL method performs better than the NJ method, studies comparing the accuracy of the ML and NJ-MCL methods are lacking. Here, accuracy of the NJ-MCL and the ML methods are examined. The concatenation approach (by progressive addition of genes) is used in a biologically realistic computer simulation to infer the accuracy of the methods. Simulation results clearly show that although NJ-MCL is computationally efficient and outperforms NJ method, the ML method is clearly much more accurate than the NJ-MCL method. The results encourage the use of the ML algorithm where datasets include up to several hundred species, but for reconstructing grand-scale phylogenies (i.e., where several thousand of taxa are included) NJ-MCL is preferred.     

机器学习结合密度泛函理论计算在材料科学中的研究进展

近年来,材料科学研究中密度泛函理论(DFT)计算与机器学习相结合的方法呈现爆炸式增长的趋势。本文综述了密度泛函理论(DFT)及其高通量方法产生的大量计算数据与机器学习相结合的原理和意义,从DFT计算的基本原理出发,重点介绍了机器学习(ML)方法的流程、常用的算法及其在催化材料预测热门研究方向中的应用,最后剖析了这个新兴领域目前存在的研究问题、挑战以及未来的发展前景。     

基于激光诱导击穿光谱(LIBS)的煤灰熔点快速检测

炉内结渣是影响火电机组和气化工艺可靠运行的关键因素之一,准确预测灰熔点可以提前调整炉膛出口温度以避免结渣。本论文采用激光诱导击穿光谱（LIBS）采集煤灰样中金属元素的光谱,分别建立煤灰中的金属元素的谱线强度与煤灰熔点的随机森林模型、支持向量机回归模型和线性回归模型,直接预测煤灰熔点温度。采用基于马氏距离(MD)的异常数据剔除算法和基于稀疏矩阵的基线估计与降噪算法（BEADS）,对粉煤灰样的全光谱数据进行了预处理。随机森林模型对粉煤灰熔点的预测平均相对误差（MRE）为54.74%,支持向量机回归模型的预测平均相对误差为60.08%,而线性回归模型的预测平均相对误差达到了9.78%。研究结果表明,线性回归模型对煤灰熔点的预测结果更准确。