Machine Learning for Sudden Cardiac Death Prediction in the Atherosclerosis Risk in Communities Study

Introduction Sudden cardiac death (SCD) is a devastating consequence often without antecedent expectation. Current risk stratification methods derived from baseline independently modeled risk factors are insufficient. Novel random forest machine learning (ML) approach incorporating time-dependent variables and complex interactions may improve SCD risk prediction. Methods Atherosclerosis Risk in Communities (ARIC) study participants were followed for adjudicated SCD. ML models were compared to standard Poisson regression models for interval data, an approximation to Cox regression, with stepwise variable selection. Eighty-two time-varying variables (demographics, lifestyle factors, clinical characteristics, biomarkers, etc.) collected at four visits over 12 years (1987-98) were used as candidate predictors. Predictive accuracy was assessed by area under the receiver operating characteristic curve (AUC) through out-of-bag prediction for ML models and 5-fold cross validation for the Poisson regression models. Results Over a median follow-up time of 23.5 years, 583 SCD events occurred among 15,661 ARIC participants (mean age 54 years and 55% women). Compared to different Poisson regression models (AUC at 6-year ranges from 0.77-0.83), the ML model improved prediction (AUC at 6-year 0.89). Top predictors identified by ML model included prior coronary heart disease (CHD), which explained 47.9% of the total phenotypic variance, diabetes mellitus, hypertension, and T wave abnormality in any of leads I, aVL, or V6. Using the top ML predictors to select variables, the Poisson regression model AUC at 6-year was 0.77 suggesting that the non-linear dependencies and interactions captured by ML, are the main reasons for its improved prediction performance. Conclusions Applying novel ML approach with time-varying predictors improves the prediction of SCD. Interactions of dynamic clinical characteristics are important for risk-stratifying SCD in the general population.

Melena Revealing an Attenuated Familial Adenomatous Polyposis in an Elderly Patient: A Case Report

We describe a patient who was diagnosed with multiple tubulleuvillous adenomas with focus of high-grade tubular dysplasia all over the colonic mucosa, discovered during a colonoscopy performed during an episode of melena. Genetic testing has identified a germline truncating mutation at the codon (5q22.2) of the adenomatous polyposis (APC) gene. This mutation is localized in the alternately spliced region of exon 12, a region which is associated with an attenuated familial adenomatous polyposis (PAFA) phenotype. Our patient had no extracolic manifestations of PAFA and none of her relatives had a history of rectocolic polyposis. Treatment consisted of colectomy with ileorectal anastomosis. PAFA is an ill-defined condition of unknown prevalence and penetrance, requiring individual treatment and lifelong monitoring. It is essential to identify these patients with a view to setting up appropriate endoscopic surveillance at an early age in family members carrying this mutation, due to the marked intra-family phenotypic variance.

Pleiotropic Effect Analysis and Marker Development for Grain Zinc and Iron Concentrations in Spring wheat

Wheat (Triticum aestivum L.) is one of the main food crops in the world and a primary source of zinc (Zn) and iron (Fe) in the human body. The genetic mechanisms underlying related traits have been clarified, thereby providing a molecular theoretical foundation for the development of germplasm resources. In this study, 23,536 high-quality DArT markers were used to map quantitative trait loci (QTL) of grain Zn (GZn) and grain Fe (GFe) concentrations in recombinant inbred lines from Avocet/Chilero. A total of 17 QTLs located on chromosomes 1BL, 2BL, 3BL, 4AL, 4BS, 5AL, 5DL, 6AS, 6BS, 6DS, and 7AS accounted for 0.38–16.62% of the phenotypic variance. QGZn.haust-4AL, QGZn.haust-7AS.1, and QGFe.haust-6BS were detected on chromosomes 4AL, 6BS, and 7AS, accounting for 10.63–16.62% of the phenotypic variance. Four stable QTLs, QGZn.haust-4AL, QGFe.haust-1BL, QGFe.haust-4AL, and QGFe.haust-5DL were located on chromosomes 1BL, 4AL, and 5DL. Three pleiotropic effects locus for GZn and GFe concentrations were located on chromosomes 1BL, 4AL, and 5DL. Two high-throughput Kompetitive Allele Specific PCR markers were developed by closely linking single nucleotide polymorphisms on chromosomes 4AL and 5DL, which were validated by a germplasm panel. Therefore, it is the most important that quantitative trait loci and KASP marker for grain zinc and iron concentrations were developed for utilizing in marker-assisted breeding and biofortification of wheat grain in breeding programs.

A Major Quantitative Trait Loci Cluster Controlling Three Components of Yield and Plant Height Identified on Chromosome 4B of Common Wheat

Wheat yield is not only affected by three components of yield, but also affected by plant height (PH). Identification and utilization of the quantitative trait loci (QTL) controlling these four traits is vitally important for breeding high-yielding wheat varieties. In this work, we conducted a QTL analysis using the recombinant inbred lines (RILs) derived from a cross between two winter wheat varieties of China, “Nongda981” (ND981) and “Nongda3097” (ND3097), exhibiting significant differences in spike number per unit area (SN), grain number per spike (GNS), thousand grain weight (TGW), and PH. A total of 11 environmentally stable QTL for these four traits were detected. Among them, four major and stable QTLs (QSn.cau-4B-1.1, QGns.cau-4B-1, QTgw.cau-4B-1.1, and QPh.cau-4B-1.2) explaining the highest phenotypic variance for SN, GNS, TGW, and PH, respectively, were mapped on the same genomic region of chromosome 4B and were considered a QTL cluster. The QTL cluster spanned a genetic distance of about 12.3 cM, corresponding to a physical distance of about 8.7 Mb. Then, the residual heterozygous line (RHL) was used for fine mapping of the QTL cluster. Finally, QSn.cau-4B-1.1, QGns.cau-4B-1, and QPh.cau-4B-1.2 were colocated to the physical interval of about 1.4 Mb containing 31 annotated high confidence genes. QTgw.cau-4B-1.1 was divided into two linked QTL with opposite effects. The elite NILs of the QTL cluster increased SN and PH by 55.71–74.82% and 14.73–23.54%, respectively, and increased GNS and TGW by 29.72–37.26% and 5.81–11.24% in two environments. Collectively, the QTL cluster for SN, GNS, TGW, and PH provides a theoretical basis for improving wheat yield, and the fine-mapping result will be beneficial for marker-assisted selection and candidate genes cloning.

Transdiagnostic phenotypes of compulsive behavior and associations with psychological, cognitive, and neurobiological affective processing

Compulsivity is a poorly understood transdiagnostic construct thought to underlie multiple disorders, including obsessive-compulsive disorder, addictions, and binge eating. Our current understanding of the causes of compulsive behavior remains primarily based on investigations into specific diagnostic categories or findings relying on one or two laboratory measures to explain complex phenotypic variance. This proof-of-concept study drew on a heterogeneous sample of community-based individuals (N = 45; 18–45 years; 25 female) exhibiting compulsive behavioral patterns in alcohol use, eating, cleaning, checking, or symmetry. Data-driven statistical modeling of multidimensional markers was utilized to identify homogeneous subtypes that were independent of traditional clinical phenomenology. Markers were based on well-defined measures of affective processing and included psychological assessment of compulsivity, behavioral avoidance, and stress, neurocognitive assessment of reward vs. punishment learning, and biological assessment of the cortisol awakening response. The neurobiological validity of the subtypes was assessed using functional magnetic resonance imaging. Statistical modeling identified three stable, distinct subtypes of compulsivity and affective processing, which we labeled “Compulsive Non-Avoidant”, “Compulsive Reactive” and “Compulsive Stressed”. They differed meaningfully on validation measures of mood, intolerance of uncertainty, and urgency. Most importantly, subtypes captured neurobiological variance on amygdala-based resting-state functional connectivity, suggesting they were valid representations of underlying neurobiology and highlighting the relevance of emotion-related brain networks in compulsive behavior. Although independent larger samples are needed to confirm the stability of subtypes, these data offer an integrated understanding of how different systems may interact in compulsive behavior and provide new considerations for guiding tailored intervention decisions.

A Saturated Map of Common Genetic Variants Associated with Human Height from 5.4 Million Individuals of Diverse Ancestries

Common SNPs are predicted to collectively explain 40-50% of phenotypic variation in human height, but identifying the specific variants and associated regions requires huge sample sizes. Here we show, using GWAS data from 5.4 million individuals of diverse ancestries, that 12,111 independent SNPs that are significantly associated with height account for nearly all of the common SNP-based heritability. These SNPs are clustered within 7,209 non-overlapping genomic segments with a median size of ~90 kb, covering ~21% of the genome. The density of independent associations varies across the genome and the regions of elevated density are enriched for biologically relevant genes. In out-of-sample estimation and prediction, the 12,111 SNPs account for 40% of phenotypic variance in European ancestry populations but only ~10%-20% in other ancestries. Effect sizes, associated regions, and gene prioritization are similar across ancestries, indicating that reduced prediction accuracy is likely explained by linkage disequilibrium and allele frequency differences within associated regions. Finally, we show that the relevant biological pathways are detectable with smaller sample sizes than needed to implicate causal genes and variants. Overall, this study, the largest GWAS to date, provides an unprecedented saturated map of specific genomic regions containing the vast majority of common height-associated variants.

Assessing the leaf shape dynamic through marker–trait association under drought stress in a rice germplasm panel

The flag leaf acts as a functional leaf in rice, Oryza sativa L., primarily supplying photosynthate to the developing grains and influencing yields to a certain extent. Drought stress damages the leaf physiology, severely affecting grain fertility. Autumn rice of northeast India is called locally as ‘ahu’ rice, and is known for its drought tolerance. Exploring diverse germplasm resources at the morphological level using an association mapping approach can aid in identifying the genomic regions influencing leaf shape dynamics. A marker–trait association (MTA) study was carried out using 95 polymorphic SSR markers and a panel of 273 ahu rice germplasm accessions in drought stress and irrigated conditions. The trials suggest that at the vegetative stage, drought stress significantly affects leaf morphology. The leaf physiology of some tolerant accessions was relatively little affected by stress and these can be considered as ideal varieties for drought conditions. The phenotypic coefficient of variance and genotypic coefficient of variance values implied moderate to high variability for the leaf traits studied. Analysis of molecular variance inferred that 11% of variation in the germplasm panel was due to differences between populations, while the remaining 89% may be attributed to a difference within subgroups formed through STRUCTURE analysis. Using the mixed linear model approach revealed 11 MTAs explaining between 4.5 and 20.0% of phenotypic variance at P > 0.001 for all the leaf traits. The study concludes that ahu rice germplasm is extremely diverse and can serve as a valuable resource for mining desirable alleles for drought tolerance.

Genetic Architecture and Candidate Genes for Pubescence Length and Density and Its Relationship With Resistance to Common Cutworm in Soybean

Soybean pubescence plays an important role in insect resistance, drought tolerance, and other stresses. Hence, a deep understanding of the molecular mechanism underlying pubescence is a prerequisite to a deeper understanding of insect resistance and drought tolerance. In the present study, quantitative trait loci (QTL) mapping of pubescence traits was performed using a high-density inter-specific linkage map of one recombinant inbred line (RIL) population, designated NJRINP. It was observed that pubescence length (PL) was negatively correlated with pubescence density (PD). A total of 10 and 9 QTLs distributed on six and five chromosomes were identified with phenotypic variance (PV) of 3.0–9.9% and 0.8–15.8% for PL and PD, respectively, out of which, eight and five were novel. Most decreased PL (8 of 10) and increased PD (8 of 9) alleles were from the wild soybean PI 342618B. Based on gene annotation, Protein ANalysis THrough Evolutionary Relationships and literature search, 21 and 12 candidate genes were identified related to PL and PD, respectively. In addition, Glyma.12G187200 from major QTLs qPL-12-1 and qPD-12-2, was identified as Ps (sparse pubescence) before, having an expression level of fivefold greater in NN 86-4 than in PI 342618B, hence it might be the candidate gene that is conferring both PL and PD. Based on gene expression and cluster analysis, three and four genes were considered as the important candidate genes of PL and PD, respectively. Besides, leaves with short and dense (SD) pubescence, which are similar to the wild soybean pubescence morphology, had the highest resistance to common cutworm (CCW) in soybean. In conclusion, the findings in the present study provide a better understanding of genetic basis and candidate genes information of PL and PD and the relationship with resistance to CCW in soybean.

Quantitative Trait Loci Mapping and Development of KASP Marker Smut Screening Assay Using High-Density Genetic Map and Bulked Segregant RNA Sequencing in Sugarcane (Saccharum spp.)

Sugarcane is one of the most important industrial crops globally. It is the second largest source of bioethanol, and a major crop for biomass-derived electricity and sugar worldwide. Smut, caused by Sporisorium scitamineum, is a major sugarcane disease in many countries, and is managed by smut-resistant varieties. In China, smut remains the single largest constraint for sugarcane production, and consequently it impacts the value of sugarcane as an energy feedstock. Quantitative trait loci (QTLs) associated with smut resistance and linked diagnostic markers are valuable tools for smut resistance breeding. Here, we developed an F1 population (192 progeny) by crossing two sugarcane varieties with contrasting smut resistance and used for genome-wide single nucleotide polymorphism (SNP) discovery and mapping, using a high-throughput genotyping method called “specific locus amplified fragment sequencing (SLAF-seq) and bulked-segregant RNA sequencing (BSR-seq). SLAF-seq generated 148,500 polymorphic SNP markers. Using SNP and previously identified SSR markers, an integrated genetic map with an average 1.96 cM marker interval was produced. With this genetic map and smut resistance scores of the F1 individuals from four crop years, 21 major QTLs were mapped, with a phenotypic variance explanation (PVE) > 8.0%. Among them, 10 QTLs were stable (repeatable) with PVEs ranging from 8.0 to 81.7%. Further, four QTLs were detected based on BSR-seq analysis. aligning major QTLs with the genome of a sugarcane progenitor Saccharum spontaneum, six markers were found co-localized. Markers located in QTLs and functional annotation of BSR-seq-derived unigenes helped identify four disease resistance candidate genes located in major QTLs. 77 SNPs from major QTLs were then converted to Kompetitive Allele-Specific PCR (KASP) markers, of which five were highly significantly linked to smut resistance. The co-localized QTLs, candidate resistance genes, and KASP markers identified in this study provide practically useful tools for marker-assisted sugarcane smut resistance breeding.

Study on fine mapping of QTL for juice yield traits of sweet sorghum (Sorghum dochna)

The stem juice yield is a key factor that influences both the biological and economic production of sweet sorghum [Sorghum dochna (Forssk.) Snowden]. To elucidate upon the genetic basis of the stem juice yield, an F₅ population developed from a cross between the low juice yielding Xinliang52 (XL52) and high juice yielding W455 lines, were used in a quantitative trait locus (QTL) analysis. A main effect of the QTL controlling stem juice yield was separated with an SSR marker called Xtxp97, which explained 46.7% of the phenotypic variance. In addition, F₅ and F₆ populations were constructed with XL52 and W452 as the parents to further verify the QTLs, and a significant correlation was found between the juice yield trait and the Xtxp97 marker. Based on the progeny tests of 29 recombinants, QJy-sbi06 was located in a region of about 21.2 kb on chromosome 6, where a candidate gene encoding an NAC transcription factor (sobic.006G147400) was identified. Combining the different population association analysis and sequencing technology showed that XL52 inserted a 1.8 kb transposon in the NAC to directly interrupt and inactivate the juice yield gene. This study also demonstrated that the colour of the leaf midribs was controlled by a single gene and was significantly positive correlated with juiciness (r = 0.784, P < 0.01). These results could lay the foundation for map-based cloning of QJy-sbi06 and provide genes or QTLs for breeding sorghum lines with a high juice yield and quality.