Evaluation of reverse transcription yield of RNA standards and forensic samples based on droplet digital PCR.

RNA analysis has shown great value in forensic science, such as body fluids and tissue identification, postmortem interval estimation, biological age prediction, etc. Currently, most RNA follow-up experiments involve reverse transcription (RT) procedures. It has been shown that the RT step is variable and has a greater impact on subsequent data analysis, especially for forensic trace samples. However, the pattern of variation between different RNA template inputs and complementary DNA (cDNA) yield is unclear. In this study, a series of 2-fold gradient dilutions of RNA standards (1 µg/µL - 0.24 ng/µL) and forensic samples (including blood samples, saliva samples, bloodstains, and saliva stains) were reverse-transcribed using EasyQuick RT MasterMix. The obtained cDNA was quantified by droplet digital PCR (ddPCR) to assess the RT yield of the ACTB gene. The results showed that the 125 ng RNA template had the highest RT yield in a 10 µL RT reaction system with the selected kit. For all stain samples, the RT yield improved as the amount of RNA template input increased since RNA quantities were below 125 ng. As many commercialized reverse transcription kits using different kinds of enzymes are available for forensic RNA research, we recommend that systematic experiments should be performed in advance to determine the amount of RNA input at the optimum RT yield when using any kit for reverse transcription experiments.

Body fluids should be identified before estimating the time since deposition (TsD) in microbiome-based stain analyses for forensics.

Identification and the time since deposition (TsD) estimation of body fluid stains from a crime scene could provide valuable information for solving the cases and are always difficult for forensics. Microbial characteristics were considered as a promising biomarker to address the issues. However, changes in the microbiota may damage the specific characteristics of body fluids. Correspondingly, incorrect body fluid identification may result in inaccurate TsD estimation. The mutual influence is not well understood and limited the codetection. In the current study, saliva, semen, vaginal secretion, and menstrual blood samples were exposed to indoor conditions and collected at eight time points (from fresh to 30 days). High-throughput sequencing based on the 16S rRNA gene was performed to characterize the microbial communities. The results showed that a longer TsD could decrease the discrimination of different body fluid stains. However, the accuracies of identification still reached a quite high value even without knowing the TsD. Correspondingly, the mean absolute error (MAE) of TsD estimation significantly increased without distinguishing the types of body fluids. The predictive TsD of menstrual blood reached a quite low MAE (1.54 ± 0.39 d). In comparison, those of saliva (6.57 ± 1.17 d), semen (6.48 ± 1.33 d), and vaginal secretion (5.35 ± 1.11 d) needed to be further improved. The great effect of individual differences on these stains limited the TsD estimation accuracy. Overall, microbial characteristics allow for codetection of body fluid identification and TsD estimation, and body fluids should be identified before estimating TsD in microbiome-based stain analyses.IMPORTANCEEmerged evidences suggest microbial characteristics could be considered a promising tool for identification and time since deposition (TsD) estimation of body fluid stains. However, the two issues should be studied together due to a potential mutual influence. The current study provides the first evidence to understand the mutual influence and determines an optimal process for codetection of identification and TsD estimation for unknown stains for forensics. In addition, we involved aged stains into our study for identification of body fluid stains, rather than only using fresh stains like previous studies. This increased the predictive accuracy. We have preliminary verified that individual differences in microbiotas limited the predictive accuracy of TsD estimation for saliva, semen, and vaginal secretion. Microbial characteristics could provide an accurate TsD estimation for menstrual blood. Our study benefits the comprehensive understanding of microbiome-based stain analyses as an essential addition to previous studies.

A robust method for simultaneous determination of eight B vitamins in human serum by liquid chromatography tandem mass spectrometry.

The level of vitamin B group in human serum is an important index of human health. Among B vitamins, cyanocobalamin in serum is unstable and its content is extremely low. Rapid and simultaneous detection of multiple B vitamins including cyanocobalamin is a challenge. Herein, we have developed a rapid and stable method that can realize the determination of thiamine, riboflavin, nicotinamide, pantothenic acid, pyridoxic acid, biotin, 5-methyltetrahydrofolate, and cyanocobalamin simultaneously in 6 min. The method was established based on protein precipitation with methanol and then chromatographic separation was achieved using Waters acquity ultra-high-performance liquid chromatography high strength silica T3 column, which was stable and sensitive especially for cyanocobalamin. Limit of quantification, precision, trueness, and matrix effect were validated according to the European Medicines Agency and United States Food and Drug guidelines and Clinical and Laboratory Standards Institute guidelines on bioanalytical method. The limit of quantification for thiamine, riboflavin, nicotinamide, pantothenic acid, pyridoxic acid, biotin, 5-methyltetrahydrofolate, and cyanocobalamin was 0.4, 0.4, 0.8, 2.0, 0.4, 0.1, 0.4, and 0.04 ng/mL separately, respectively. Intra- and interday precisions were 1.1%-12.4% and 2.0%-13.5%, respectively. The relative errors were between 0.3% and 13.3%, and the matrix effects were between 2.6% and 10.4%.

Rapid simultaneous determination of 7 fat-soluble vitamins in human serum by ultra high performance liquid chromatography tandem mass spectrometry.

Accurate detection of vitamins is critically important for clinical diagnosis, metabolomics and epidemiological studies. However, the amounts of different vitamins vary dramatically in human serum. It is a challenge to achieve simultaneous detection of multiple vitamins rapidly. Herein, we developed and validated a sensitive and specific method using ultra high-performance liquid chromatography tandem mass spectrometry (UPLC-MS/MS) for simultaneous quantification of 7 fat-soluble vitamins (FSVs) across their physiological concentrations in serum for the first time, which was subjected to protein precipitation, liquid-liquid extraction to an organic phase, evaporation to dryness and reconstitution with acetonitrile. In the present procedure, retinol (vitamin A), ergocalciferol (25-OH-D2), cholecalciferol (25-OH-D3), α-tocopherol (vitamin E), phylloquinone (vitamin K1), menatetrenone-4 (MK-4), and menaquinone-7 (MK-7) were detected in one analytical procedure for the first time within 5.0 min by triple quadrupole tandem mass spectrometry. The limit of quantification (LOQ) for vitamin A was 10.0 ng mL-1, LOQs for 25-OH-D2 and 25-OH-D3 were 1.0 ng mL-1, LOQ for vitamin E was 100.0 ng mL-1, and LOQs for vitamin K1, MK-4 and MK-7 were 0.10 ng mL-1, respectively, with a correlation (R2) of 0.995-0.999. Recoveries ranged from 80.5% to 118.5% and the intra-day and inter-day coefficients of variance (CVs) were 0.72-8.89% and 3.2-9.0% respectively. The method was validated according to the European Medicines Agency (EMA) and U.S. Food and Drug guidelines and C62-A on bioanalytical methods, and was used for clinical routine determination.

Rapid and visual detection of specific bacteria for saliva and vaginal fluid identification with the lateral flow dipstick strategy.

Identification of body fluids is critical for crime scene reconstruction, and a source of investigation source of investigative leads. In recent years, microbial DNA analysis using sequencing and quantitative real-time polymerase chain reaction have been used to identify body fluids. However, these techniques are time-consuming, expensive, and require complex workflows. In this study, a new method for simultaneous detection of Streptococcus salivarius and Lactobacillus crispatus using polymerase chain reaction (PCR) in combination with a lateral flow dipstick (LFD) was developed to identify saliva and vaginal fluid in forensic samples. LFD results can be observed with the naked eye within 3 min with a sensitivity of 0.001 ng/µL DNA. The PCR-LFD assay was successfully used to detect S. salivarius and L. crispatus in saliva and vaginal fluid respectively, and showed negative results in blood, semen, nasal fluid, and skin. Moreover, saliva and vaginal fluid were detectable even at an extremely high mixing ratio of sample DNA (1:999). Saliva and vaginal fluid were identified in various mock forensic samples. These results indicate that saliva and vaginal fluid can be effectively detected by identifying S. salivarius and L. crispatus, respectively. Furthermore, we have shown that DNA samples used to identify saliva and vaginal fluid can also provide a complete short tandem repeat (STR) profile when used as source material for forensic STR profiling. In summary, our results suggest that PCR-LFD is a promising assay for rapid, simple, reliable, and efficient identification of body fluids.

Visible-light-mediated catalyst-free synthesis of unnatural α-amino acids and peptide macrocycles.

The visible light induced, photocatalysts or photoabsorbing EDA complexes mediated cleavage of pyridinium C-N bond were reported in the past years. Here, we report an ionic compound promote homolytic cleavage of pyridinium C-N bond by exploiting the photonic energy from visible light. This finding is successfully applied in deaminative hydroalkylation of a series of alkenes including naturally occurring dehydroalanine, which provides an efficient way to prepare ß-alkyl substituted unnatural amino acids under mild and photocatalyst-free conditions. Importantly, by using this protocol, the deaminative cyclization of peptide backbone N-terminals is realized. Furthermore, the use of Et3N or PPh3 as reductants and H2O as hydrogen atom source is a practical advantage. We anticipate that our protocol will be useful in peptide synthesis and modern peptide drug discovery.

Visible Light Induced Cu-Catalyzed Asymmetric C(sp3)-H Alkylation.

The asymmetric functionalization of C-H is one of the most attractive strategies in asymmetric synthesis. In the past decades, catalytic enantioselective C(sp3)-H functionalization has been intensively studied and successfully applied in various asymmetric bond formations, whereas asymmetric C(sp3)-H alkylation was not well developed. Photoredox catalysis has recently emerged as an efficient way to synthesize organic compounds under mild conditions. Despite many photoinduced stereoselective reactions that have been achieved, the related enantioselective C(sp3)-C(sp3) coupling is challenging, especially of the photocatalytic asymmetric C(sp3)-H radical alkylation. Here, we report a visible light induced Cu catalyzed asymmetric sp3 C-H alkylation, which is effective for coupling with unbiased primary, secondary, and tertiary alkyl fragments in high enantioselectivities. This reaction would provide a new approach for the synthesis of important molecules such as unnatural α-amino acids and late-stage functionalization of bioactive compounds, and will be useful for modern peptide synthesis and drug discovery.

LGAN: Lung segmentation in CT scans using generative adversarial network.

Lung segmentation in Computerized Tomography (CT) images plays an important role in various lung disease diagnosis. Most of the current lung segmentation approaches are performed through a series of procedures with manually empirical parameter adjustments in each step. Pursuing an automatic segmentation method with fewer steps, we propose a novel deep learning Generative Adversarial Network (GAN)-based lung segmentation schema, which we denote as LGAN. The proposed schema can be generalized to different kinds of neural networks for lung segmentation in CT images. We evaluated the proposed LGAN schema on datasets including Lung Image Database Consortium image collection (LIDC-IDRI) and Quantitative Imaging Network (QIN) collection with two metrics: segmentation quality and shape similarity. Also, we compared our work with current state-of-the-art methods. The experimental results demonstrated that the proposed LGAN schema can be used as a promising tool for automatic lung segmentation due to its simplified procedure as well as its improved performance and efficiency.

Visible-light promoted regioselective amination and alkylation of remote C(sp3)-H bonds.

The C-N cross coupling reaction has always been a fundamental task in organic synthesis. However, the direct use of N-H group of aryl amines to generate N-centered radicals which would couple with alkyl radicals to construct C-N bonds is still rare. Here we report a visible light-promoted C-N radical cross coupling for regioselective amination of remote C(sp3)-H bonds. Under visible light irradiation, the N-H groups of aryl amines are converted to N-centered radicals, and are then trapped by alkyl radicals, which are generated from Hofmann-Löffler-Freytag (HLF) type 1,5-hydrogen atom transfer (1,5-HAT). With the same strategy, the regioselective C(sp3)-C(sp3) cross coupling is also realized by using alkyl Hantzsch esters (or nitrile) as radical alkylation reagents. Notably, the α-C(sp3)-H of tertiary amines can be directly alkylated to form the C(sp3)-C(sp3) bonds via C(sp3)-H - C(sp3)-H cross coupling through the same photoredox pathway.

3D-GLCM CNN: A 3-Dimensional Gray-Level Co-Occurrence Matrix-Based CNN Model for Polyp Classification via CT Colonography.

Accurately classifying colorectal polyps, or differentiating malignant from benign ones, has a significant clinical impact on early detection and identifying optimal treatment of colorectal cancer. Convolution neural network (CNN) has shown great potential in recognizing different objects (e.g. human faces) from multiple slice (or color) images, a task similar to the polyp differentiation, given a large learning database. This study explores the potential of CNN learning from multiple slice (or feature) images to differentiate malignant from benign polyps from a relatively small database with pathological ground truth, including 32 malignant and 31 benign polyps represented by volumetric computed tomographic (CT) images. The feature image in this investigation is the gray-level co-occurrence matrix (GLCM). For each volumetric polyp, there are 13 GLCMs, computed from each of the 13 directions through the polyp volume. For comparison purpose, the CNN learning is also applied to the multi-slice CT images of the volumetric polyps. The comparison study is further extended to include Random Forest (RF) classification of the Haralick texture features (derived from the GLCMs). From the relatively small database, this study achieved scores of 0.91/0.93 (two-fold/leave-one-out evaluations) AUC (area under curve of the receiver operating characteristics) by using the CNN on the GLCMs, while the RF reached 0.84/0.86 AUC on the Haralick features and the CNN rendered 0.79/0.80 AUC on the multiple-slice CT images. The presented CNN learning from the GLCMs can relieve the challenge associated with relatively small database, improve the classification performance over the CNN on the raw CT images and the RF on the Haralick features, and have the potential to perform the clinical task of differentiating malignant from benign polyps with pathological ground truth.

Expert knowledge-infused deep learning for automatic lung nodule detection.

BACKGROUND: Computer aided detection (CADe) of pulmonary nodules from computed tomography (CT) is crucial for early diagnosis of lung cancer. Self-learned features obtained by training datasets via deep learning have facilitated CADe of the nodules. However, the complexity of CT lung images renders a challenge of extracting effective features by self-learning only. This condition is exacerbated for limited size of datasets. On the other hand, the engineered features have been widely studied. OBJECTIVE: We proposed a novel nodule CADe which aims to relieve the challenge by the use of available engineered features to prevent convolution neural networks (CNN) from overfitting under dataset limitation and reduce the running-time complexity of self-learning. METHODS: The CADe methodology infuses adequately the engineered features, particularly texture features, into the deep learning process. RESULTS: The methodology was validated on 208 patients with at least one juxta-pleural nodule from the public LIDC-IDRI database. Results demonstrated that the methodology achieves a sensitivity of 88% with 1.9 false positives per scan and a sensitivity of 94.01% with 4.01 false positives per scan. CONCLUSIONS: The methodology shows high performance compared with the state-of-the-art results, in terms of accuracy and efficiency, from both existing CNN-based approaches and engineered feature-based classifications.

Improved computer-aided detection of pulmonary nodules via deep learning in the sinogram domain.

Computer aided detection (CADe) of pulmonary nodules plays an important role in assisting radiologists' diagnosis and alleviating interpretation burden for lung cancer. Current CADe systems, aiming at simulating radiologists' examination procedure, are built upon computer tomography (CT) images with feature extraction for detection and diagnosis. Human visual perception in CT image is reconstructed from sinogram, which is the original raw data acquired from CT scanner. In this work, different from the conventional image based CADe system, we propose a novel sinogram based CADe system in which the full projection information is used to explore additional effective features of nodules in the sinogram domain. Facing the challenges of limited research in this concept and unknown effective features in the sinogram domain, we design a new CADe system that utilizes the self-learning power of the convolutional neural network to learn and extract effective features from sinogram. The proposed system was validated on 208 patient cases from the publicly available online Lung Image Database Consortium database, with each case having at least one juxtapleural nodule annotation. Experimental results demonstrated that our proposed method obtained a value of 0.91 of the area under the curve (AUC) of receiver operating characteristic based on sinogram alone, comparing to 0.89 based on CT image alone. Moreover, a combination of sinogram and CT image could further improve the value of AUC to 0.92. This study indicates that pulmonary nodule detection in the sinogram domain is feasible with deep learning.

Speciation of inorganic arsenic in a sequential injection dual mini-column system coupled with hydride generation atomic fluorescence spectrometry.

The separation and speciation of inorganic arsenic(III) and arsenic(V) are facilitated by employing a novel sequential injection system incorporating two mini-columns followed by detection with hydride generation atomic fluorescence spectrometry. An octadecyl immobilized silica mini-column is used for selective retention of the complex between As(III) and APDC, while the sorption of As(V) is readily accomplished by a 717 anion exchange resin mini-column. The retained As(III)-PDC complex and As(V) are effectively eluted with a 3.0 mol L(-1) hydrochloric acid solution as stripping reagent, which well facilitates the ensuing hydride generation process via reaction with tetrahydroborate. With a sampling volume of 1.0 mL and an eluent volume of 100 microL for both species, linear ranges of 0.05-1.5 microg L(-1) for As(III) and 0.1-1.5 microg L(-1) for As(V) are obtained, along with enrichment factors of 7.0 and 8.2, respectively. Precisions of 2.8% for As(III) and 2.9% for As(V) are derived at the concentration level of 1.0 microg L(-1). The practical applicability of the procedure has been demonstrated by analyzing a certified reference material of riverine water (SLRS-4), in addition to spiking recovery in a lake water sample matrix.