Toward amino acid typing for proteins in FFLUX

Continuing the development of the FFLUX, a multipolar polarizable force field driven by machine learning, we present a modern approach to atom‐typing and building transferable models for predicting atomic properties in proteins. Amino acid atomic charges in a peptide chain respond to the substitution of a neighboring residue and this response can be categorized in a manner similar to atom‐typing. Using a machine learning method called kriging, we are able to build predictive models for an atom that is defined, not only by its local environment, but also by its neighboring residues, for a minimal additional computational cost. We found that prediction errors were up to 11 times lower when using a model specific to the correct group of neighboring residues, with a mean prediction of ∼0.0015 au. This finding suggests that atoms in a force field should be defined by more than just their immediate atomic neighbors. When comparing an atom in a single alanine to an analogous atom in a deca‐alanine helix, the mean difference in charge is 0.026 au. Meanwhile, the same difference between a trialanine and a deca‐alanine helix is only 0.012 au. When compared to deca‐alanine models, the transferable models are up to 20 times faster to train, and require significantly less ab initio calculation, providing a practical route to modeling large biological systems. © 2016 The Authors. Journal of Computational Chemistry Published by Wiley Periodicals, Inc.     

Automated and efficient quantum chemical determination and energetic ranking of molecular protonation sites

We present an automated quantum chemical protocol for the determination of preferred protonation sites in organic and organometallic molecules containing up to a few hundred atoms. It is based on the Foster–Boys orbital localization method, whereby we automatically identify lone pairs and π orbitals as possible protonation sites. The method becomes efficient in conjunction with the robust and fast GFN‐xTB semiempirical method proposed recently (Grimme et al ., J. Chem. Theory Comput . 2017, 13 , 1989). The protonated isomers that are found within a few seconds to minutes of computational wall‐time on a standard desktop computer are then energetically refined using density functional theory (DFT), where we use a high‐level double‐hybrid reference method to benchmark GFN‐xTB and low‐cost DFT approaches. The proposed DFT/GFN‐xTB/LMO composite protocol is generally applicable to almost arbitrary molecules including transition metal complexes. Importantly it is found that even in electronically complicated cases, the GFN‐xTB optimized protomer structures are reasonable and can safely be used in single‐point DFT calculations. Corrections from energy to free energy mostly have a small effect on computed protomer populations. The resulting protomer equilibrium is valuable, for example, in the context of electrospray ionization mass spectrometry where it may help identify the ionized species and assist the interpretation of the experiment. © 2017 Wiley Periodicals, Inc.     

Accurate van der Waals force field for gas adsorption in porous materials

An accurate van der Waals force field (VDW FF) was derived from highly precise quantum mechanical (QM) calculations. Small molecular clusters were used to explore van der Waals interactions between gas molecules and porous materials. The parameters of the accurate van der Waals force field were determined by QM calculations. To validate the force field, the prediction results from the VDW FF were compared with standard FFs, such as UFF, Dreiding, Pcff, and Compass. The results from the VDW FF were in excellent agreement with the experimental measurements. This force field can be applied to the prediction of the gas density (H₂, CO₂, C₂H₄, CH₄, N₂, O₂) and adsorption performance inside porous materials, such as covalent organic frameworks (COFs), zeolites and metal organic frameworks (MOFs), consisting of H, B, N, C, O, S, Si, Al, Zn, Mg, Ni, and Co. This work provides a solid basis for studying gas adsorption in porous materials. © 2017 Wiley Periodicals, Inc.     

富镉离子CdS量子点电化学发光法检测L-半脐氨酸

采用一锅法,通过控制镉硫比合成了表面富镉离子的硫化镉量子点,利用L-半胱氨酸可与量子点表面Cd2+结合,使量子点表面钝化,从而增强其电化学发光信号的性质,实现了对L-半胱氨酸的选择性检测.对合成的量子点进行了表征,优化了检测条件.在优化的条件下,L-半胱氨酸在5.0×10-9~1.0×10-5 mol/L浓度范围内与ECL信号呈良好的线性关系,检出限为1.2×10-9 mol/L(S/N=3).本方法对L-半胱氨酸具有良好的选择性,用于实际样品中L-半胱氨酸的测定,结果令人满意.     

基于Mn掺杂ZnS量子点的室温磷光传感应用的研究进展

与一般有机染料分子相比,半导体材料量子点具有优异的光学性能,在多个领域得到了广泛的应用.量子点具有窄而对称且可调的发射波长、宽激发强吸收、抗光漂白能力强以及水溶性好等诸多优势,引起了研究者广泛关注.为了增加量子点的斯托克斯位移从而很好地避免量子点的自猝灭现象,引入掺杂物是一种很有效的方式.掺杂量子点不仅保留了量子点原有的优点,而且还赋予量子点额外的优异性能.如Mn掺杂ZnS量子点生物相容性好,不含Cd和Hg等有害元素,而且Mn2+的加入使其具有优异的室温磷光特性.磷光检测能很好地避开生物背景荧光的干扰,使得Mn掺杂ZnS量子点能够广泛应用于磷光生物分析.本文综述了Mn掺杂ZnS量子点在室温磷光分析中的研究进展,着重介绍了几种具有启发意义的设计策略,包括其发光机理以及应用于离子、分子以及生物大分子等的检测.     

基于稀土离子介导石墨烯量子点荧光开关的凝血酶生物传感器

稀土离子(Er3+)可与荧光石墨烯量子点(GQDs)表面的含氧基团发生配位,在Er3+介导下形成高配位数的GQDs/Er3+配合物,引起GQDs聚集而使其荧光减弱.凝血酶(Tb)中的氮和氧等原子可与Er3+发生配位作用,从而与GQDs竞争结合Er3+,减弱了GQDs与Er3+的作用而使其荧光恢复.通过检测GQDs的荧光即可实现对Tb活性的高灵敏分析,构建了基于Er3+介导GQDs荧光开关的Tb传感方法,采用透射电镜、原子力显微镜、红外吸收光谱以及荧光光谱等对传感机理进行了研究.本方法对Tb的检出限低至0.049 nmol/L,其它蛋白质对Tb检测无明显干扰,实际样品中Tb加标回收率为98.0%~105.3%,相对标准偏差为0.6%~4.2%.     

Ionic impurities in nematic liquid crystal doped with quantum dots CdSe/ZnS

We investigated the dielectric losses and the ionic currents in the nematic liquid crystal (NLC) doped with semiconductor quantum dots (QDs) of CdSe/ZnS core – shell type and covered with trioctylphosphine oxide (TOPO) molecules. The dielectric loss tangent of the NLC composites increased with increasing the QDs concentration from 0.1 to 0.3 wt%. The density of mobile ions in the composites increased linearly and the average values of ions mobility in the composites decreased with increasing the QDs concentration. The fast ions with the mobility of about 10^–10 m²/V·s and the slow ions with the mobility of about 10^–11 m²/V·s were detected in the NLC composites. The growth of the content of slow ions took place with increasing the QDs concentrations. Increasing the dielectric loss tangent was observed with increasing the duration of sonication time of the NLC composites to prepare homogeneous suspensions. The fragmentation of the CdS/ZnS shell as a result of the sonication may lead to the appearance of the slow ions in the NLC composites.     

三(3,5-二氟苄基)氯化锡和四(邻氯苄基)锡的合成、晶体结构及量子化学

3,5-二氟苄基氯和邻氯基苄基氯在适当的溶剂中与锡粉反应,合成了三(3,5-二氟苄基)氯化锡(1)和四(邻氯苄基)锡(2),经X射线衍射方法测定了新化合物的晶体结构。化合物1属单斜晶系,空间群为C2/c,晶体学参数:a=1.858 33(11)nm,b=1.140 98(7)nm,c=2.690 06(16)nm,β=109.288(10)°,V=5.383 6(6)nm3,Z=8,Dc=1.532 g·cm~(-3),μ(Mo Kα)=13.61 cm~(-1),F(000)=2 480,R1=0.085 1,wR~2=0.168 1。化合物2属单斜晶系,空间群为P21/m,晶体学参数:a=0.585 54(5)nm,b=1.969 74(18)nm,c=0.857 86(8)nm,β=95.204 0(10)°,V=0.985 34(15)nm3,Z=2,Dc=1.805 g·cm~(-3),μ(Mo Kα)=14.91 cm-1,F(000)=524,R1=0.054 0,wR_2=0.163 9;中心锡原子为畸变四面体构型。对其结构进行量子化学从头计算,探讨了化合物的稳定性、分子轨道能量以及部分前沿分子轨道的组成特征。测定了化合物的热稳定性。     

Analysis of temperature and stress field in Al alloy’s twin wire welding

By analyzing the shape of twin wire welding’s arcs and the track of droplets’ transition, the phenomenon that the twin wire welding’s fore arc and rear arc all deflect to the middle of the two arcs is found. Based on this the double ellipsoid heat source model is amended, and a heat source model which can apply to calculate the twin wire welding’s temperature field is put forward. This model is testified by actual experiment of temperature sampling. By comparing the temperature field of twin wire welding and single wire welding, the results show that twin wire welding has slender weld pool the end part of which is ellipsoid, and its HAZ is narrower than that of single wire welding. So, twin wire welding can not only reduce the Al alloy generating hot crack, but can also weaken the “overaging” softened phenomenon of heat treated strengthening Al alloy. In the end, the evolving rules of 2219 Al alloy’s longitudinal and transverse stress when welded with twin wire welding are analyzed.