Comparison of Four Machine Learning Techniques for Prediction of Intensive Care Unit Length of Stay in Heart Transplantation Patients.

Background: Post-operative heart transplantation patients often require admission to an intensive care unit (ICU). Early prediction of the ICU length of stay (ICU-LOS) of these patients is of great significance and can guide treatment while reducing the mortality rate among patients. However, conventional linear models have tended to perform worse than non-linear models. Materials and Methods: We collected the clinical data of 365 patients from Wuhan Union Hospital who underwent heart transplantation surgery between April 2017 and August 2020. The patients were randomly divided into training data (N = 256) and test data (N = 109) groups. 84 clinical features were collected for each patient. Features were validated using the Least Absolute Shrinkage and Selection Operator (LASSO) regression's fivefold cross-validation method. We obtained Shapley Additive explanations (SHAP) values by executing package "shap" to interpret model predictions. Four machine learning models and logistic regression algorithms were developed. The area under the receiver operating characteristic curve (AUC-ROC) was used to compare the prediction performance of different models. Finally, for the convenience of clinicians, an online web-server was established and can be freely accessed via the website https://wuhanunion.shinyapps.io/PredictICUStay/. Results: In this study, 365 consecutive patients undergoing heart transplantation surgery for moderate (NYHA grade 3) or severe (NYHA grade 4) heart failure were collected in Wuhan Union Hospital from 2017 to 2020. The median age of the recipient patients was 47.2 years, while the median age of the donors was 35.58 years. 330 (90.4%) of the donor patients were men, and the average surgery duration was 260.06 min. Among this cohort, 47 (12.9%) had renal complications, 25 (6.8%) had hepatic complications, 11 (3%) had undergone chest re-exploration and 19 (5.2%) had undergone extracorporeal membrane oxygenation (ECMO). The following six important clinical features were selected using LASSO regression, and according to the result of SHAP, the rank of importance was (1) the use of extracorporeal membrane oxygenation (ECMO); (2) donor age; (3) the use of an intra-aortic balloon pump (IABP); (4) length of surgery; (5) high creatinine (Cr); and (6) the use of continuous renal replacement therapy (CRRT). The eXtreme Gradient Boosting (XGBoost) algorithm presented significantly better predictive performance (AUC-ROC = 0.88) than other models [Accuracy: 0.87; sensitivity: 0.98; specificity: 0.51; positive predictive value (PPV): 0.86; negative predictive value (NPV): 0.93]. Conclusion: Using the XGBoost classifier with heart transplantation patients can provide an accurate prediction of ICU-LOS, which will not only improve the accuracy of clinical decision-making but also contribute to the allocation and management of medical resources; it is also a real-world example of precision medicine in hospitals.

[Research Advance of the Roles of N6-methyladenosine in the Regulation of Hematopoiesis--Review].

There are more than 150 types of chemical modifications in RNA, mainly methylation, which are widely distributed in all kinds of RNA, including messenger RNA, transfer RNA, ribosomal RNA, non-coding small RNA and long non-coding RNA. In recent years, the identification of RNA methylation modification enzymes and the development of high-throughput sequencing technology at transcriptome level laid a foundation for revealing the expression and function of genes regulated by chemical modification of RNA. In this review, the most recent advances of RNA methylation, especially N6-methyladenosine (m6a) in the blood system, including the regulation of RNA methyltransferases, RNA demethylases and RNA binding proteins on normal and malignant hematopoiesis through the regulation of RNA methylation level were summarized briefly.

The long-non-coding RNA NEAT1 is a novel target for Alzheimer's disease progression via miR-124/BACE1 axis.

OBJECTIVES: Long-non-coding RNAs (lncRNAs) have been involved in central nervous system recently. A number of studies have reported that lncRNA NEAT1 exerts critical roles in neurodegenerative disorder. Beta-site amyloid precursor protein cleaving enzyme 1 (BACE1) has been reported to exert function in the accumulation of amyloid-ß (Aß). Moreover, BACE1 acts as a target of miR-124 in the progression of AD. So far, the biological role and underlying mechanisms of NEAT1 and miR-124 in AD remains elusive. METHODS: The relative NEAT1 and miR-124 expression was examined by qRT-PCR in the tissues and cells line of AD. Cell apoptosis was examined by FACS. Luciferase reporter assay was performed to verify that miR-124 is a direct target of NEAT1, and BACE1 is a downstream target of miR-124. qRT-PCR and western blot analysis were also performed to determinate the BACE1 and the phosphorylation of tau protein. RESULTS: NEAT1 was notably up-regulated and miR-124 was remarkably down-regulated in AD mouse model. Knockdown of NEAT1 or overexpression of miR-124 showed the protective effects on cellular AD model induced by Aß. Moreover, miR-124 expression could be up- and down-regulated by suppression or overexpression of NEAT1, respectively. In addition, the expression of BACE1 was the potential functional target of miR-124. These findings suggested that NEAT1 might play a vital role in the development of AD by regulating miR-124/BACE1 axis. DISCUSSION: The present study showed that NEAT1 worked as a regulating factor to promote the development of AD via modulating miR-124/BACE1 axis, which might be considered as a novel target in AD treatment.

Allele-specific expression and alternative splicing in horse×donkey and cattle×yak hybrids.

Divergence of gene expression and alternative splicing is a crucial driving force in the evolution of species; to date, however the molecular mechanism remains unclear. Hybrids of closely related species provide a suitable model to analyze allele-specific expression (ASE) and allele-specific alternative splicing (ASS). Analysis of ASE and ASS can uncover the differences in cis-regulatory elements between closely related species, while eliminating interference of trans-regulatory elements. Here, we provide a detailed characterization of ASE and ASS from 19 and 10 transcriptome datasets across five tissues from reciprocal-cross hybrids of horse×donkey (mule/hinny) and cattle×yak (dzo), respectively. Results showed that 4.8%-8.7% and 10.8%-16.7% of genes exhibited ASE and ASS, respectively. Notably, lncRNAs and pseudogenes were more likely to show ASE than protein-coding genes. In addition, genes showing ASE and ASS in mule/hinny were found to be involved in the regulation of muscle strength, whereas those of dzo were involved in high-altitude adaptation. In conclusion, our study demonstrated that exploration of genes showing ASE and ASS in hybrids of closely related species is feasible for species evolution research.

Hydroxycinnamoylmalated flavone C-glycosides from Lemna japonica.

Three previously undescribed flavone C-glycosides (1-3), along with seven known ones (4-10), were isolated and characterized from the smallest flowering aquatic plant, Lemna japonica. On the basis of spectroscopic analysis and alkaline hydrolysis, compounds 1-3 were identified to be luteolin 6-C-(2″-O-trans-caffeoyl-d-malate)-ß-glucoside (1), apigenin 6-C-(2″-O-trans-caffeoyl-d-malate)-ß-glucoside (2), and luteolin 6-C-(2″-O-trans-coumaroyl-d-malate)-ß-glucoside (3). Compounds 1-3 are characteristic of a trans-coumaroyl-d-malate or trans-caffeoyl-d-malate linked to C-2″ of the glucose, which was reported for the first time. Compounds 1-3 exhibited weak cytotoxicity against HepG-2, SW-620, and A-549 cell lines, with IC50 values between 42.5 and 19.2µg/ml, and moderate antioxidant activity. Meanwhile compound 3 displayed moderate nematocidal activity with an EC50 value of 1.56mg/ml.

A Dy4 single-molecule magnet and its Gd(iii), Tb(iii), Ho(iii), and Er(iii) analogues encapsulated by an 8-hydroxyquinoline Schiff base derivative and ß-diketonate coligand.

Five new tetranuclear complexes based on an 8-hydroxyquinoline Schiff base derivative and the ß-diketone coligand, [Ln4(acac)4L6(µ3-OH)2]·CH3CN·0.5CH2Cl2 (Ln = Gd (1), Tb (2), Dy (3), Ho (4) and Er (5); HL = 5-(benzylidene)amino-8-hydroxyquinoline; acac = acetylacetonate) have been synthesized, and structurally and magnetically characterized. Complexes 1-5 have similar tetranuclear structures. Each LnIII ion is eight coordinated and its coordination polyhedra can be described as being in a distorted square-antiprismatic geometry. The magnetic studies reveal that 1 features the magnetocaloric effect (MCE) with the magnetic entropy change of -ΔSm(T) = 25.08 J kg-1 K-1 at 2 K for ΔH = 7 T, and 3 displays the slow magnetic relaxation behavior of Single Molecule Magnets (SMMs) with the anisotropic barrier of 86.20 K and the pre-exponential factor τ0 = 2.99 × 10-8 s.

Modulation of the relaxation dynamics of linear-shaped tetranuclear rare-earth clusters through utilizing different solvents.

Nine new tetranuclear centrosymmetric linear complexes, [RE4(dbm)8L2(DMF)2]·nCH2Cl2·mC2H3N (RE = Y (1), Tb (2), Dy (3), Ho (4), Er (5), Lu (6)) and [RE4(dbm)8L2(C2H5OH)2]·nCH3CN (RE = Tb (7), Dy (8), Ho (9)) (HL = 2-[(2-(hydroxyimino)propanehydrazide)methyl]-8-hydroxyquinoline and dbm = 1,3-diphenyl-1,3-propanedione) have been synthesized. Complexes 1-9 are tetranuclear complexes. Magnetic studies reveal that both DyIII-based complexes (3 and 8) exhibit single-molecule magnet (SMM) behavior under a zero dc field. Furthermore, complex 3 presents one relaxation process under a zero dc field, while application of an external dc field (1500 Oe) induces multi-relaxation signals of the ac magnetic susceptibility.