Search

BackgroundNeonatal screening for inherited metabolic diseases (IMDs) has been revolutionized by tandem mass spectrometry (MS/MS). This study aimed to enhance neonatal screening for IMDs using machine learning (ML) techniques.MethodsThe study involved the analysis of a comprehensive dataset comprising 309,102 neonatal screening records collected in the Ningbo region, China. An advanced ML system model, encompassing nine distinct algorithms, was employed for the purpose of predicting the presence of 31 different IMDs. The model was compared with traditional cutoff schemes to assess its diagnostic efficacy. Additionally, 180 suspected positive cases underwent further evaluation.ResultsThe ML system exhibited a significantly reduced positive rate, from 1.17% to 0.33%, compared to cutoff schemes in the initial screening, minimizing unnecessary recalls and associated stress. In suspected positive cases, the ML system identified 142 true positives with high sensitivity (93.42%) and improved specificity (78.57%) compared to the cutoff scheme. While false negatives emerged, particularly in heterozygous carriers, our study revealed the potential of the ML system to detect asymptomatic cases.ConclusionThis research provides valuable insights into the potential of ML in pediatric medicine for IMD diagnosis through neonatal screening, emphasizing the need for accurate carrier detection and further research in this domain.

Phenylketonuria (PKU) is a genetic disorder with amino acid metabolic defect, which does great harms to the development of newborns and children. Early diagnosis and treatment can effectively prevent the disease progression. Here we developed a PKU screening model using random forest classifier (RFC) to improve PKU screening performance with excellent sensitivity, false positive rate (FPR) and positive predictive value (PPV) in all the validation dataset and two testing Chinese populations. RFC represented outstanding advantages comparing several different classification models based on machine learning and the traditional logistic regression model. RFC is promising to be applied to neonatal PKU screening.

Xinkeshu tablets (XKST), a traditional Chinese patent medicine (CPM), have served in the clinical treatment of cardiovascular diseases (CVDs) for decades. However, its pharmacodyamic material basis was still unclear, and the holistic quality control has not been well established due to the lack of systematic research on the quality markers. In this experiment, the heart rate recovery rate of a zebrafish larva was used to evaluate the traditional pharmacological effect of XKST i.e., antiarrhythmic effect. The HPLC fingerprints of 16 batches of XKST samples were obtained, and antiarrhythmic components of XKST were identified by establishing the spectrum-effect relationship between HPLC fingerprints and heart rate recovery rate of zebrafish larva with orthogonal signal correction and partial least squares regression (OSC–PLSR) analysis. The anticardiovascular disease components of XKST were identified by mapping the targets related to CVDs in network pharmacology. The compounds of XKST absorbed and exposed in vivo were identified by ultra-high performance liquid chromatography Q-Exactive high-resolution mass spectrometry (UHPLC-Q-Exactive HRMS). Based on the earlier studies, combined with five principles for identifying quality markers and verified by a zebrafish arrhythmia model, danshensu, salvianolic acid A, salvianolic acid B, daidzein, and puerarin were identified as quality markers of XKST. In total, 16 batches of XKST samples were further quantified with the method established in this study. Our study laid the foundation for the quality control of XKST. The integrated strategy used in the study of XKST could be applied for the identification and quantification of quality markers of other CPMs as well.

Given the labor-consuming nature of model establishment, model transfer has become a considerable topic in the study of near-infrared (NIR) spectroscopy. Recently, many new algorithms have been proposed for the model transfer of spectra collected by the same types of instruments under different situations. However, in a practical scenario, we need to deal with model transfer between different types of instruments. To expand model applicability, we must develop a method that could transfer spectra acquired from different types of NIR spectrometers with different wavenumbers or absorbance. Therefore, in our study, we propose a new methodology based on improved principal component analysis (IPCA) for calibration transfer between different types of spectrometers. We adopted three datasets for method evaluation, including public pharmaceutical tablets (dataset 1), corn data (dataset 2), and the spectra of eight batches of samples acquired from the plasma ethanol precipitation process collected by FT-NIR and MicroNIR spectrometers (dataset 3). In the calibration transfer for public datasets, IPCA displayed comparable results with the classical calibration transfer method using piecewise direct standardization (PDS), indicating its obvious ability to transfer spectra collected from the same types of instruments. However, in the calibration transfer for dataset 3, our proposed IPCA method achieved a successful bi-transfer between the spectra acquired from the benchtop and micro-instruments with/without wavelength region selection. Furthermore, our proposed method enabled improvements in prediction ability rather than the degradation of the models built with original micro spectra. Therefore, our proposed method has no limitations on the spectrum for model transfer between different types of NIR instruments, thus allowing a wide application range, which could provide a supporting technology for the practical application of NIR spectroscopy.

Primary carnitine deficiency (PCD) is an autosomal recessive disorder that could result in sudden death. It is caused by a defect in the carnitine transporter encoded by SLC22A5 (Solute Carrier Family 22 Member 5, MIM:603377). Currently, a number of variants in SLC22A5 have been identified, however, the PCD prevalence and its variants in Ningbo area are unclear. In this study, we screened 265,524 newborns by using tandem mass spectrometry. Variants in SLC22A5 were further detected by next-generation sequencing in individuals with abnormal free carnitine levels (C0). We identified 53 newborns with abnormal C0 levels and 26 with variants in SLC22A5. Among them, 16 with compound heterozygous or homozygous variants in SLC22A5 were diagnosed with PCD, suggesting the PCD birth prevalence in Ningbo city was 1/16,595. Moreover, the C0 level was significantly (P = 0.013) higher in PCD patients than in those with one variant. Besides, the c.1400C > G (p. S467C) and c.51C > G (p. F17L) variants were the most frequent and six novel variants are all predicted to be damaging. This study reports the largest PCD patients in Ningbo area by newborn screening and expands the variant spectrum of SLC22A5. Our findings demonstrate the clinical value of combining NBS program results with DNA analysis for the diagnosis of PCD.

Cervical discogenic pain (CDP) is mainly induced by cervical disc degeneration. However, how CDP modulates the functional interactions within the pain network remains unclear. In the current study, we studied the changed resting-state functional connectivities of pain network with 40 CDP patients and 40 age-, gender-matched healthy controls. We first defined the pain network with the seeds of the posterior insula (PI). Then, whole brain and seed-to-target functional connectivity analyses were performed to identify the differences in functional connectivity between CDP and healthy controls. Finally, correlation analyses were applied to reveal the associations between functional connectivities and clinical measures. Whole-brain functional connectivity analyses of PI identified increased functional connectivity between PI and thalamus (THA) and decreased functional connectivity between PI and middle cingulate cortex (MCC) in CDP patients. Functional connectivity analyses within the pain network further revealed increased functional connectivities between bilateral PI and bilateral THA, and decreased functional connectivities between left PI and MCC, between left postcentral gyrus (PoCG) and MCC in CDP patients. Moreover, we found that the functional connectivities between right PI and left THA, between left PoCG and MCC were negatively and positively correlated with the visual analog scale, respectively. Our findings provide direct evidence of how CDP modulates the pain network, which may facilitate understanding of the neural basis of CDP.

Cervical discogenic pain (CDP) is a clinically common pain syndrome caused by cervical disk degeneration. A large number of studies have reported that CDP results in brain functional impairments. However, the detailed dynamic brain functional abnormalities in CDP are still unclear. In this study, using resting-state functional magnetic resonance imaging, we explored the neural basis of CDP with 40 CDP patients and 40 age-, gender-matched healthy controls to delineate the changes of the voxel-level static and dynamic amplitude of low frequency fluctuations (ALFF). We found increased static ALFF in left insula (INS) and posterior precuneus (PCu), and decreased static ALFF in left precentral/postcentral gyrus (PreCG/PoCG), thalamus (THA), and subgenual anterior cingulate cortex in CPD patients compared to healthy controls. We also found decreased dynamic ALFF in left PreCG/PoCG, right posterior middle temporal gyrus, and bilateral THA. Moreover, we found that static ALFF in left PreCG/PoCG and dynamic ALFF in THA were significantly negatively correlated with visual analog scale and disease duration, respectively. Our findings provide the neurophysiological basis for CDP and facilitate understanding the neuropathology of CDP.

A three-liquid-phase (TLP) salting-out extraction system composed of n-hexane/ethanol/ sodium carbonate/water was investigated to extract oils, saponins, proteins and polysaccharides simultaneously from waste liquor of processing sea cucumber. The effects of the ratio of ethanol to sodium carbonate, n-hexane concentration and extraction time were investigated. The results showed that 86.7% of oils were distributed in the top n-hexane phase, 82.9% of saponins in the middle ethanol phase, 93.2% of proteins and 92.9% of polysaccharides in the interface between the middle and the bottom salt phase when the system composed of 28% (w/w) n-hexane/11.52% (w/w) ethanol/8.64% (w/w) sodium carbonate was used at 37 °C for 0.5 h. When the system was progressively enlarged from 30 g to 25 kg, the yield of polysaccharides, proteins, oils and saponins was decreased only by 1.0%, 2.8%, 1.0% and 2.6%, respectively. The recycle of salt and solvents in three-liquid-phase system was also studied, and the results showed that the recovery of n-hexane, ethanol and salt were 81.4%, 80.8% and 72.0%, respectively. Recycling materials for the extraction, the yield for proteins and oils decreased by 2.0% and 5.4%, respectively, comparing with the pure system. [ABSTRACT FROM AUTHOR]