Machine learning models to predict in-hospital mortality in septic patients with diabetes.

Background: Sepsis is a leading cause of morbidity and mortality in hospitalized patients. Up to now, there are no well-established longitudinal networks from molecular mechanisms to clinical phenotypes in sepsis. Adding to the problem, about one of the five patients presented with diabetes. For this subgroup, management is difficult, and prognosis is difficult to evaluate. Methods: From the three databases, a total of 7,001 patients were enrolled on the basis of sepsis-3 standard and diabetes diagnosis. Input variable selection is based on the result of correlation analysis in a handpicking way, and 53 variables were left. A total of 5,727 records were collected from Medical Information Mart for Intensive Care database and randomly split into a training set and an internal validation set at a ratio of 7:3. Then, logistic regression with lasso regularization, Bayes logistic regression, decision tree, random forest, and XGBoost were conducted to build the predictive model by using training set. Then, the models were tested by the internal validation set. The data from eICU Collaborative Research Database (n = 815) and dtChina critical care database (n = 459) were used to test the model performance as the external validation set. Results: In the internal validation set, the accuracy values of logistic regression with lasso regularization, Bayes logistic regression, decision tree, random forest, and XGBoost were 0.878, 0.883, 0.865, 0.883, and 0.882, respectively. Likewise, in the external validation set 1, lasso regularization = 0.879, Bayes logistic regression = 0.877, decision tree = 0.865, random forest = 0.886, and XGBoost = 0.875. In the external validation set 2, lasso regularization = 0.715, Bayes logistic regression = 0.745, decision tree = 0.763, random forest = 0.760, and XGBoost = 0.699. Conclusion: The top three models for internal validation set were Bayes logistic regression, random forest, and XGBoost, whereas the top three models for external validation set 1 were random forest, logistic regression, and Bayes logistic regression. In addition, the top three models for the external validation set 2 were decision tree, random forest, and Bayes logistic regression. Random forest model performed well with the training and three validation sets. The most important features are age, albumin, and lactate.

Effect of iron supplementation on platelet count in adult patients with iron deficiency anemia.

Iron deficiency anemia (IDA) affects more than 1.2 billion individuals globally. In addition to anemia, reactive thrombocytosis is also a common clinical hematological condition in patients with IDA. However, some case reports have described the thrombotic complications in association with IDA-induced thrombocytosis. Patients with a high risk of thrombosis need prompt identification and effective treatment to prevent thrombotic complications. While iron replacement treatment has been shown to decrease platelet count in this context, there is limited published evidence on how iron supplementation affects the thrombocytosis caused by IDA. We retrospectively examined the clinical records of 440 patients with IDA from an RCT completed from 1 January 2016, to 30 December 2017, and data obtained from this study was used for post hoc analysis to examine the effect of iron on platelet count in IDA-induced thrombocytosis.The mean ± standard deviation (SD) platelet counts of the 440 patients with IDA was 310.23 ± 98.72 × 109/L. With baseline platelet counts>450 × 109 /L as the cutoff for thrombocytosis, patients were divided into 2 groups: 36 (8.1%) in the IDA with thrombocytosis group (mean ± SD platelet count, 521.67 ± 73.85 × 109/L) and the remaining 404 in the IDA without thrombocytosis group (mean ± SD platelet count, 291.39 ± 76.11 × 109/L).Differences were found in baseline characteristics including white blood cell (WBC) count, hemoglobin (Hb) level, mean corpuscular volume (MCV), transferrin saturation (TSAT), serum iron (SI) level, and total iron-binding capacity (TIBC) between the two groups (P < .05). From baseline to 8 weeks of continuous iron supplementation treatment, the mean platelet counts in both groups were decreased at 2-week treatment intervals. And in the IDA with thrombocytosis group, half of the patients resolved thrombocytosis after 2 weeks of iron supplementation, and the counts of all patients with thrombocytosis decreased below 450 × 109 /L within 6 weeks.In conclusion, the rate of reactive thrombocytosis in patients with IDA was 8.1%. IDA patients with thrombocytosis showed more severe anemia, lower ferritin, and more advanced iron deficiency than those without thrombocytosis. Platelet counts of half of the patients with thrombocytosis reduced below cut off of 450 × 109/L for thrombocytosis after 2 weeks of treatment, and all patients resolved thrombocytosis after 6 weeks. Our study provided clinical evidence for more effective and individualized iron management in the future. IDA patients with thrombocytosis should take active iron treatment and increase follow-up frequency to prevent thrombotic events. For patients with persistent thrombocytosis, a concomitant clonal process should be considered.

Improving Hypoxia Adaption Causes Distinct Effects on Growth and Bioactive Compounds Synthesis in an Entomopathogenic Fungus Cordyceps militaris.

Cordyceps militaris is an entomopathogenic fungus producing a variety of bioactive compounds. To meet the huge demand for medicinal and edible products, industrialized fermentation of mycelia and cultivation of stromata have been widely developed in China. The content of bioactive metabolites of C. militaris, such as cordycepin, is higher when cultivated on silkworm pupae than on rice or in broth. However, compared with other cultivation methods, C. militaris grows more slowly and accumulates less biomass. The hypoxic environment in pupa hemocoel is one of environmental factor which is not existed in other cultivation methods. It is suggested that hypoxia plays an important role on the growth and the synthesis of bioactive compounds in C. militaris. Here, we demonstrated that the distinct effects on the growth and synthesis of bioactive compounds employing different strategies of improving hypoxia adaption. The introduction of Vitreoscilla hemoglobin enhanced growth, biomass accumulation, and crude polysaccharides content of C. militaris. However, cordycepin production was decreased to 9-15% of the control group. Meanwhile, the yield of adenosine was increased significantly. Nonetheless, when the predicted bHLH transcription factor of sterol regulatory element binding proteins (SREBPs) was overexpressed in C. militaris to improve the hypoxia adaption of fungal cells, cordycepin content was significantly increased more than two-fold. These findings reveal the role of SREBPs on growth and bioactive compounds synthesis. And it also provides a scientific basis for rationally engineering strains and optimization strategies of air supply in cultivation and fermentation.

[Optimizing of cDNA preparation for next generation sequencing].

Next generation sequencing has already been used for genomic analysis of microorganis, human being, animals, and plants. Sample preparation is prerequisite and most important for large-scale sequencing. There are two major interferences for large-scale sequencing, polyA and abundant genes' concealment for rare genes. In order to solve these problems, we used total RNA extracted from violaceae leaves to produce double stranded cDNA. DSN nuclease was used to treat the ds cDNA prior to removing the polyA. Randomly sequencing 100 clones of the treated cDNA showed that there were 94 independent clones in the treated sample, and the sequences did not contained polyA. However, only 62 independent clones were found in the untreated sample, and 15 of the sequencing files were affected by polyA. By randomly sequencing of the treated cDNA, we also found two clones encoded two interested genes. We failed to isolate these genes although the protein mass peaks of them had been found in the MALDI-TOF trace. Furthermore, we designed primers from two known genes with different expression abundances. The PCR yields were approaching similar using the treated cDNAs as templates. These results showed that, removal of the polyA and enrichment of rare genes with DSN can meet the requirements of large-scale sequencing and discovery of new genes.

[Exploration on the relationship between pathogenesis for yin-deficiency of acute myocardial infarction and immediate prognosis as well as its mechanism].

OBJECTIVE: To study the relationship between pathogenesis for Yin-deficiency of acute myocardial infarction (AMI) and immediate prognosis as well as its neuro-endocrine mechanism. METHODS: According to the TCM standard of Syndrome Differentiation of Deficiency Syndrome, 194 patients with AMI were classified into the typical Yin-deficiency group (n = 26), the non-typical Yin-deficiency group (n = 61) and the non-Yin-deficiency group (n = 107). Their venous blood samples were collected in the morning while lying on their backs to detect plasma levels of adrenaline, noradrenaline, aldosterone, atrial natriuretic peptide (ANP), corticoid and myocardial enzymes, as well as their hospitalization days and mortality in hospitalized period were calculated and compared in the three groups, with the 30 healthy persons as control. RESULTS: Levels of serum creatine phosphokinase, creatine phosphokinase isozyme, plasma adrenaline, noradrenaline, aldosterone, hospitalization days and mortality were higher in the Yin-deficiency groups than in the non-Yin-deficiency group (P < 0.05, P < 0.01). As compared the indexes between the typical and the non-typical Yin-deficiency groups, significant difference only showed in plasma aldosterone and ANP, which was significantly higher in the former (P < 0.05, P < 0.01). Plasma corticoid level was insignificantly different between the Yin-deficieny groups. CONCLUSION: Patients with AMI of Yin-deficiency type was severer in myocardial damage, with longer hospitalization period and higher mortality, it is probably due to the hyper-activated sympathetic-adrenaline system and strengthened activity of aldosterone in them.