[A deep-learning model for the assessment of coronary heart disease and related risk factors via the evaluation of retinal fundus photographs]

Objective: To develop and validate a deep learning model based on fundus photos for the identification of coronary heart disease (CHD) and associated risk factors. Methods: Subjects aged>18 years with complete clinical examination data from 149 hospitals and medical examination centers in China were included in this retrospective study. Two radiologists, who were not aware of the study design, independently evaluated the coronary angiography images of each subject to make CHD diagnosis. A deep learning model using convolutional neural networks (CNN) was used to label the fundus images according to the presence or absence of CHD, and the model was proportionally divided into training and test sets for model training. The prediction performance of the model was evaluated in the test set using monocular and binocular fundus images respectively. Prediction efficacy of the algorithm for cardiovascular risk factors (e.g., age, systolic blood pressure, gender) and coronary events were evaluated by regression analysis using the area under the receiver operating characteristic curve (AUC) and R² correlation coefficient. Results: The study retrospectively collected 51 765 fundus images from 25 222 subjects, including 10 255 patients with CHD, and there were 14 419 male subjects in this cohort. Of these, 46 603 fundus images from 22 701 subjects were included in the training set and 5 162 fundus images from 2 521 subjects were included in the test set. In the test set, the deep learning model could accurately predict patients' age with an R² value of 0.931 (95%CI 0.929-0.933) for monocular photos and 0.938 (95%CI 0.936-0.940) for binocular photos. The AUC values for sex identification from single eye and binocular retinal fundus images were 0.983 (95%CI 0.982-0.984) and 0.988 (95%CI 0.987-0.989), respectively. The AUC value of the model was 0.876 (95%CI 0.874-0.877) with either monocular fundus photographs and AUC value was 0.885 (95%CI 0.884-0.888) with binocular fundus photographs to predict CHD, the sensitivity of the model was 0.894 and specificity was 0.755 with accuracy of 0.714 using binocular fundus photographs for the prediction of CHD. Conclusion: The deep learning model based on fundus photographs performs well in identifying coronary heart disease and assessing related risk factors such as age and sex.

Identification and characterization of the 14-3-3 gene family in switchgrass

Members of the 14-3-3 family of proteins are conserved regulatory proteins that are widely found in eukaryotes and play crucial roles in diverse physiological processes, including responses to different stresses. Although genome-wide analysis of 14-3-3 proteins has been performed in a few plant species, it has not been performed in switchgrass. In this study, we identified 21 switchgrass 14-3-3 proteins (designated PvGF14a to PvGF14u) and examined genes for improved stress tolerance in this species. A phylogenetic tree was constructed to demonstrate that PvGF14 proteins can be divided into six groups, and that PvGF14 proteins belonging to each class exhibit similar gene structure. A phylogenetic analysis of PvGF14 proteins among switchgrass, Arabidopsis, and rice was conducted. Ten PvGF14 proteins were found to be orthologous to several abiotic stresses, and these were particularly responsive proteins in Arabidopsis and rice. Tissue-specific expression profiles showed that PvGF14a, PvGF14k, PvGF14l, and PvGF14m may play significant roles in the regulation of lignin metabolism, and that PvGF14r may participate in flower development. Taken together, these data suggest that PvGF14 proteins may be involved in various biosynthesis.

Orthogonal design in the optimization of a start codon targeted (SCoT) PCR system in Roegneria kamoji Ohwi

Roegneria kamoji Ohwi is an excellent forage grass due to its high feeding value and high resistance to some biotic and abiotic stresses. However, the start codon targeted (SCoT) polymorphism has not been conducted on R. kamoji. In this study, an orthogonal L₁₆ (4⁵) design was employed to investigate the effects of five factors (Mg²⁺, dNTPs, Taq DNA polymerase, primer, and template DNA) on the polymerase chain reaction (PCR) to determine the optimal SCoT-PCR system for R. kamoji. The results showed that the most suitable conditions for SCoT-PCR in R. kamoji included 1.5 mM Mg²⁺, 0.15 mM dNTPs, 1.0 U Taq DNA polymerase, 0.4 pM primer, and 40 ng template DNA. SCoT primers 39 and 41 were used to verify the stability of the optimal reaction system, and amplification bands obtained from diverse samples were found to be clear, rich, and stable in polymorphisms, indicating that this reaction system can be used for SCoT-PCR analysis of R. kamoji. We have developed a simple and rapid way to study the mutual effects of factors and to obtain positive results through the use of an orthogonal design L₁₆ (4⁵) to optimize the SCoT-PCR system. This method may provide basic information for molecular marker-assisted breeding and analyses of genetic diversity in R. kamoji.

Optimization of SCoT-PCR reaction system in Dactylis glomerata by orthogonal design

The effects of 5 factors (template DNA, Mg(2+), dNTPs, Taq DNA polymerase, and primer) on the polymerase chain reaction (PCR) were investigated to optimize the start codon targeted polymor-phism (SCoT)-PCR system of Dactylis glomerata L., using an orthogo-nal design L16 (4(5)). A suitable SCoT-PCR system for D. glomerata was established; the 20 μL reaction volume contained 3.0 mM Mg(2+), 0.2 mM dNTPs, 1.0 U Taq DNA polymerase, 0.2 μM primer, 20 ng tem-plate DNA, and 2 μL 10X buffer. Each factor had a different effect on the amplification reaction, and the concentration of dNTPs had the larg-est effect on the SCoT-PCR system. We tested 10 orchardgrass samples to determine and verify the stability of the reaction system. The results showed that amplified bands from diverse materials were clear, stable, and rich in polymorphisms, indicating that the optimized system was very stable.

Assessment of genetic diversity of bermudagrass germplasm from southwest China and Africa by using AFLP markers

Cynodon dactylon (L.) Pers. var. dactylon (common bermudagrass) is widely distributed geographically between approximately 45°N and 45°S latitude, penetrating to approximately 53°N latitude in Europe. The extensive variation of morphological and adaptive characteristics of the taxon has been substantially documented, but information is lacking on DNA molecular variation in geographically disparate forms. The genetic diversity of 51 wild accessions of bermudagrass from southwest China (Sichuan, Chongqing, Yunnan, Guizhou, and Tibet) and 8 African bermudagrass was analyzed using amplified fragment length polymorphism molecular markers. A total of 670 polymorphic bands were detected with 11 primer combinations, of which 663 (98.74%) bands were found to be polymorphic. The genetic similarity among the accessions ranged from 0.64-0.96 with an average of 0.78. All 59 wild accessions were clustered into 5 eco-geographic groups, and nearly all accessions from the same area were classified into the same group and were found to be associated with their geographical distributions. Therefore, complex geographical and ecological environments are important factors for the genetic structure and geographical distribution of C. dactylon.

Genetic diversity and relationships in cultivars of Lolium multiflorum Lam. using sequence-related amplified polymorphism markers

Sequence-related amplified polymorphism (SRAP) markers were used to analyze and estimate the genetic variability, level of diversity, and relationships among 20 cultivars and strains of annual ryegrass (Lolium multiflorum Lam.). Eighteen SRAP primer combinations generated 334 amplification bands, of which 298 were polymorphic. The polymorphism information content ranged from 0.4715 (me10 + em1) to 0.5000 (me5 + em7), with an average of 0.4921. The genetic similarity coefficient ranged from 0.4304 to 0.8529, and coefficients between 0.65 and 0.90 accounted for 90.00%. The cluster analysis separated the accessions into five groups partly according to their germplasm resource origins.

Genetic diversity and relationship of chicory (Cichorium intybus L.) using sequence-related amplified polymorphism markers

Chicory is a crop with economically important roles and is cultivated worldwide. The genetic diversity and relationship of 80 accessions of chicories and endives were evaluated by sequence-related amplified polymorphism (SRAP) markers to provide a theoretical basis for future breeding programs in China. The polymorphic rate was 96.83%, and the average polymorphic information content was 0.323, suggesting the rich genetic diversity of chicory. The genetic diversity degree of chicory was higher (GS = 0.677) than that of endive (GS = 0.701). The accessions with the highest genetic diversity (effective number of alleles, NE = 1.609; Nei's genetic diversity, H = 0.372; Shannon information index, I = 0.556) were from Italy. The richest genetic diversity was revealed in a chicory line (NE = 1.478, H = 0.289, I = 0.443) among the 3 types (line, wild, and cultivar). The chicory genetic structure of 8 geographical groups showed that the genetic differentiation coefficient (GST) was 14.20% and the number of immigrants per generation (Nm) was 3.020. A GST of 6.80% and an Nm of 6.853 were obtained from different types. This observation suggests that these chicory lines, especially those from the Mediterranean region, have potential for providing rich genetic resources for further breeding programs, that the chicory genetic structure among different countries obviously differs with a certain amount of gene flow, and that SRAP markers could be applied to analyze genetic relationships and classifications of Cichorium intybus and C. endivia.

Genetic diversity and relationships among accessions of five crested wheatgrass species (Poaceae: Agropyron) based on gliadin analysis

Agropyron Gaertn. is the most important genus in Triticeae (Poaceae), which includes many forage grasses with high economic value. The genetic diversity and relationships of 36 accessions from five crested wheatgrass species were analyzed by gliadin markers. A total of 54 product bands were detected after acid polyacrylamide gel electrophoresis (A-PAGE), of which 100% were polymorphic. The genetic similarity coefficient based on Nei-Li's method ranged from 0.065 to 0.755 with an average of 0.451. The Shannon diversity information index showed that there was a high level of genetic diversity among the accessions. An unweighted pair group method with arithmetic average (UPGMA) dendrogram was constructed based on the Nei-Li's genetic similarity coefficients, which showed the phylogenetic relationships among accessions of different species. Analysis of molecular variance (AMOVA) showed that the proportion of variance explained by inter- and intraspecific variance was 9.34 and 90.66%, respectively, which revealed that the genetic variations within species were higher than the variations among species. Based on pairwise genetic distances (ΦST) among species, the cluster analysis indicated that A. mongolicum had a low-affinity relationship with other species, while A. fragile showed a close relationship with A. cristatum ssp pectinatum. Finally, the implications of the results for the taxonomy of Agropyron were discussed.

Genetic variation and comparison of orchardgrass (Dactylis glomerata L.) cultivars and wild accessions as revealed by SSR markers

Orchardgrass is a highly variable, perennial forage grass that is cultivated throughout temperate and subtropical regions of the world. Despite its economic importance, the genetic relationship and distance among and within cultivars are largely unknown but would be of great interest for breeding programs. We investigated the molecular variation and structure of cultivar populations, compared the level of genetic diversity among cultivars (Baoxing, Anba, Bote, and Kaimo), subspecies (Dactylis glomerata ssp Woronowii) and advanced breeding line (YA02-116) to determine whether there is still sufficient genetic diversity within presently used cultivars for future breeding progress in China. Twenty individuals were analyzed from each of six accessions using SSR markers; 114 easily scored bands were generated from 15 SSR primer pairs, with an average of 7.6 alleles per locus. The polymorphic rate was 100% among the 120 individuals, reflecting a high degree of genetic diversity. Among the six accessions, the highest genetic diversity was observed in Kaimo (H = 0.2518; I = 0.3916; P = 87.3%) and 02-116 had a lower level of genetic diversity (H = 0.1806; I = 0.2788; P = 58.73%) compared with other cultivars tested. An of molecular variance revealed a much larger genetic variation within accessions (65%) than between them (35%). This observation suggests that these cultivars have potential for providing rich genetic resource for further breeding program. Furthermore, the study also indicated that Chinese orchardgrass breeding has involved strong selection for adaptation to forage production, which may result in restricted genetic base of orchardgrass cultivar.

Evaluation of a statistical model for the formation of poly [acryloyl hydroxyethyl starch] microspheres

PURPOSE

To characterize the network structure of Poly(Acryloyl Hydroxyethyl Starch) (Ac-HES) microspheres and test the theoretical model and the hypothesis that the rate of swelling of microspheres is inversely related to the extent of crosslinking.

METHODS

Microspheres were prepared with varying degrees of derivatization (DD) and molar ratios (MR) and subjected to the characterization of matrix structure by dynamic and equilibrium swelling studies utilizing direct microscopic observation and the Flory-Rehner equation. The dependence of average molecular weight between crosslinking Mc, on DD and MR were compared to test the validity of the model.

RESULTS

Study of the dependence of Mc on the microspheres preparation parameters, DD and MR, showed that at constant MR, the Mc decreased with DD, while at constant DD, the Mc initially decreased with MR to a minimum, and then increased with MR, complying with the model prediction. Dynamic swelling of microspheres showed a monotonical increase to equilibrium size featured by two time variables, Tp and Teq, that were dependent on Mc; this permitted a conceptual view of the general structure of the Ac-HES microspheres. The Mc, which was more accurately determined by the weight method (as opposed to volume method), was independent of the size of microspheres although there was evidence of variation among particles within a batch.

CONCLUSIONS

The results validated the model in describing the polymerization/crosslinking reaction of the Ac-HES microsphere system and suggested that Mc is the principal factor in controlling release.

Development of a statistical model for the formation of poly [acryloyl hydroxyethyl starch] microspheres

PURPOSE

To develop a mathematical model for predicting the molecular weight between crosslinks, Mc, of poly[Acryloyl Hydroxyethyl Starch] (Ac-HES) microspheres system and to identify and evaluate the key microsphere preparation parameters which affect the Mc of the formed microsphere structure based on the developed model.

METHODS

Link probability generating functions (LPGFs) based on the classical branching theory were used to derive a model for the calculation of Mc for the Ac-HES system. Based on the developed model, simulation was made to study the effects of the microsphere preparation variables on Mc of the formed microspheres. The process variables were the degree of derivatization (DD) of the Ac-HES, the molar ratio (MR) of the Ac-HES to acrylamide monomer, the fractional conversion of the unsaturation (alpha), the initiator efficiency (f), the molar concentration of initiator (I), the fraction of intramolecular cyclization (c), and the total weight of the reactable monomer and polymer (s).

RESULTS

A model to describe the crosslinking reaction of Ac-HES system and predict Mc was developed. Simulation based on the model showed that Mc decreased as alpha increased and reached a limiting value before total conversion. At constant alpha, Mc initially decreased with MR to a minimum and then increased with MR; while Mc decreased monotonically with DD. I and c affected Mc only at very low alpha and changes in s and f had no effect on Mc.

CONCLUSIONS

Simulation based on the model suggested that the most important microsphere preparation parameters influencing Mc of the Ac-HES system are the number of functional groups on the Ac-HES (DD) and the stoichiometry (MR) of the crosslinking reaction.

Measured conversion efficiencies of P45, paraterphenyl, tetraphenyl butadiene, and sodium salicylate phosphors in the soft-x-ray wavelength range

The measured conversion efficiencies at 9.89, 23.6, 44.7, and 160 Å of P45 phosphor screens, as well as those of paraterphenyl, tetraphenyl butadiene, and sodium salicylate at 9.89, 44.7, and 67.6 Å, are presented. The conversion efficiency is defined as the ratio of photoelectrons ejected from the photocathode of a visible detector, which are excited by the scintillated photons that are emitted from the phosphor in a solid angle of 2π, to the number of soft-x-ray photons incident on the phosphor. The effect of the phosphor's thickness on the conversion efficiency was studied. The P45 phosphor converts the soft-x-ray photon (10-200 Å) into an order of magnitude more visible photons than the low-Z phosphors. The P45 phosphor screen used in conjunction with a photomultiplier tube offers a soft-x-ray photodetector with a conversion efficiency that ranges from 0.5 at 160 Å to 12 at 9.89 Å and a high electronic gain.

Photosystem II function and herbicide binding sites during photoinhibition of spinach chloroplasts in-vivo and in-vitro

The time courses of some Photosystem II (PS II) parameters have been monitored during in-vivo and in-vitro photoinhibition of spinach chloroplasts, at room temperature and at 10 °C or 0 °C. Exposing leaf discs of low-light grown spinach at 25 °C to high light led to photoinhibition of chloroplasts in-vivo as manifested by a parallel decrease in the number of functional PS II centres, the variable chlorophyll fluorescence at 77K (F v /F m ), and the number of atrazine-binding sites. When the photoinhibitory treatment was given at 10 °C, the former two parameters declined in parallel but the loss of atrazine-binding sites occurred more slowly and to a lesser extent. During in-vitro photoinhibition of chloroplast thylakoids at 25 °C, the loss of functional PS II centres proceeded slightly more rapidly than the loss of atrazine-binding sites, and this difference in rate was further increased when the thylakoids were photoinhibited at 0 °C. During the recovery phase of leaf discs (up to 9 h) the increases in F v /F m preceded that of the number of functional PS II centres, while only a further decline in the number of atrazine-binding sites was observed. The recovery of variable chlorophyll fluorescence and the concentration of functional PS II centres occurred more rapidly at 25 °C than at 10 °C. These results suggest that the photoinhibition of PS II function is a relatively temperature-independent early photochemical event, whereas the changes in the concentration of herbicide-binding sites appear to be a more complex biochemical process which can occur with a delayed time course.