Integrating Machine Learning into Statistical Methods in Disease Risk Prediction Modeling: A Systematic Review.

Background: Disease prediction models often use statistical methods or machine learning, both with their own corresponding application scenarios, raising the risk of errors when used alone. Integrating machine learning into statistical methods may yield robust prediction models. This systematic review aims to comprehensively assess current development of global disease prediction integration models. Methods: PubMed, EMbase, Web of Science, CNKI, VIP, WanFang, and SinoMed databases were searched to collect studies on prediction models integrating machine learning into statistical methods from database inception to 2023 May 1. Information including basic characteristics of studies, integrating approaches, application scenarios, modeling details, and model performance was extracted. Results: A total of 20 eligible studies in English and 1 in Chinese were included. Five studies concentrated on diagnostic models, while 16 studies concentrated on predicting disease occurrence or prognosis. Integrating strategies of classification models included majority voting, weighted voting, stacking, and model selection (when statistical methods and machine learning disagreed). Regression models adopted strategies including simple statistics, weighted statistics, and stacking. AUROC of integration models surpassed 0.75 and performed better than statistical methods and machine learning in most studies. Stacking was used for situations with >100 predictors and needed relatively larger amount of training data. Conclusion: Research on integrating machine learning into statistical methods in prediction models remains limited, but some studies have exhibited great potential that integration models outperform single models. This study provides insights for the selection of integration methods for different scenarios. Future research could emphasize on the improvement and validation of integrating strategies.

Characterization of the FLA Gene Family in Tomato (Solanum lycopersicum L.) and the Expression Analysis of SlFLAs in Response to Hormone and Abiotic Stresses.

Fasciclin-like arabinogalactan proteins (FLAs), a subclass of arabinogalactan proteins (AGPs), participate in mediating plant growth, development, and response to abiotic stress. However, the characterization and function of FLAs in tomato are currently unknown. In this study, members of the tomato FLA family are characterized and analyzed in relation to their response to phytohormonal and abiotic stresses. The results show that a total of 24 FLA members were characterized in tomato. The structural domain analysis showed that these members have a high protein similarity. The expression profiles of different tissues indicated that the genes of most members of the tomato FLA gene family are highly expressed in roots, but to a lower extent in fruits. qRT-PCR analysis revealed that all 24 tomato FLA genes are responsive to ABA and MeJA. SlFLAs showed a positive response to salt and cold stress. SlFLA1, SlFLA12, and SlFLA14 are significantly induced under darkness. SlFLA1 and SlFLA3 are significantly induced under drought stress. This study provides a basis for a further understanding of the role of tomato FLA homologous genes in plant response to abiotic stress and lays the foundation for further research on the function of FLAs in tomato.

Chromosome-level genome assembly and population genetic analysis of a near-threatened rosewood species (Dalbergia cultrata Pierre Graham ex Benth) provide insights into its evolutionary and cold stress responses.

Dalbergia cultrata Pierre Graham ex Benth (D. cultrata) is a precious rosewood tree species that grows in the tropical and subtropical regions of Asia. In this study, we used PacBio long-reading sequencing technology and Hi-C assistance to sequence and assemble the reference genome of D. cultrata. We generated 171.47 Gb PacBio long reads and 72.43 Gb Hi-C data and yielded an assembly of 10 pseudochromosomes with a total size of 690.99 Mb and Scaffold N50 of 65.76 Mb. The analysis of specific genes revealed that the triterpenoids represented by lupeol may play an important role in D. cultrata's potential medicinal value. Using the new reference genome, we analyzed the resequencing of 19 Dalbergia accessions and found that D. cultrata and D. cochinchinensis have the latest genetic relationship. Transcriptome sequencing of D. cultrata leaves grown under cold stress revealed that MYB transcription factor and E3 ubiquitin ligase may be playing an important role in the cold response of D. cultrata. Genome resources and identified genetic variation, especially those genes related to the biosynthesis of phytochemicals and cold stress response, will be helpful for the introduction, domestication, utilization, and further breeding of Dalbergia species.

Fasudil may alleviate alcohol-induced astrocyte damage by modifying lipid metabolism, as determined by metabonomics analysis.

Alcohol dependence is a chronic, relapsing encephalopathy characterized by compulsive craving for alcohol, loss of control over alcohol use, and the presence of negative emotions and physical discomfort when alcohol is unavailable. Harmful use of alcohol is one of the greatest risk factors for death, illness, and disability. Rho kinase inhibitors have neuroprotective effects. This study used metabonomics analysis to assess untreated astrocytes, astrocytes exposed to 75 mmol/L of alcohol, and astrocytes exposed to 75 mmol/L of alcohol and treated with 15 µg/mL fasudil for 24 h. One of the clearest differences between the alcohol-exposed and fasudil-treated alcohol-exposed groups was the abundance of lipids and lipid-like molecules, although glycerophospholipid metabolism was comparable in both groups. Our findings show that fasudil may alleviate alcohol-induced astrocyte damage by modifying lipid metabolism, providing a new approach for preventing and treating alcohol dependence.

Resequencing of Rosa rugosa accessions revealed the history of population dynamics, breed origin, and domestication pathways.

BACKGROUND: Rosa rugosa is a shrub that originated in China and has economic and ecological value. However, during the development of R. rugosa, the genetic background was chaotic, and the genetic structure among different wild populations was unclear, as well as wild and cultivated accessions. Here, we report whole-genome resequencing of wild and cultivated R. rugosa accessions. RESULTS: A total of 19,041,284 SNPs were identified in 188 R. rugosa accessions and 3 R. chinensis accessions by resequencing. Population genetic analysis revealed that cultivated and wild groups were separated very early. All R. rugosa accessions were divided into 8 categories based on genetic structure: (1) Weihai, Yantai, and Liaoning category, (2) Jilin category, and (3) Hammonasset category (above three are wild); (4) traditional varieties, (5) hybrids between R. rugosa and R. chinensis, (6) Zizhi Rose, (7) Kushui Rose, (8) hybrids between R. rugosa and R. multiflora. We found that the heterozygosity and genetic diversity of wild accessions were generally lower than those of cultivated individuals. The genes that were selected during cultivation were identified, and it was found that these genes were mainly related to environmental adaptation and growth. CONCLUSIONS: The Jilin population was the oldest population and later migrated to Liaoning and then migrated to Yantai and Weihai by sea regression in the Bohai Basin. The Hammonasset naturalized population probably originated from the Jilin population and then experienced separate differentiation. The long-term asexual reproduction pattern of R. rugosa decreased genetic diversity in the wild population. During R. rugosa cultivation, the ancestors of the Jilin population were involved in breeding traditional varieties, after which almost no wild individuals were engaged in breeding. However, in recent decades, cross breeding of R. rugosa started the utilization of wild germplasms. In comparison, some other species play important roles in variety formation. Few genes related to economic traits were selected, suggesting no directional domestication in the R. rugosa cultivation process.

The Akebia Genus as a Novel Forest Crop: A Review of Its Genetic Resources, Nutritional Components, Biosynthesis, and Biological Studies.

The genus Akebia belongs to the Lardizabalaceae family and comprises five species that are primarily distributed in East Asia. Plants of the Akebia genus comprise deciduous and semi-evergreen perennial twining vines that have been used in Chinese herbal medicine for at least 2000 years. The plants of this genus have the potential to form a novel forest crop with high nutritional and economic value because their fruit has a delicious sweet taste and rich nutrient components. In this study, we organized, analyzed, and evaluated the available published scientific literature on the botanical, ecological, and phytochemical characteristics of Akebia plants. Based on these studies, we briefly introduced botanical and ecological characteristics and focused on reviewing the development and utilization of wild genetic resources in the genus Akebia. We further explored the genus' rich nutritional components, such as triterpenes, flavonoids, polyphenols, polysaccharides, and fatty acids, and their potential use in food and health improvement applications. In addition, several papers describing advances in biotechnological research focusing on micropropagation, nutrient biosynthesis, and fruit ripeness were also included. This review provides comprehensive knowledge of the Akebia genus as a new forest crop for food and fruit utilization, and we also discuss future breeding and research prospects.

Characterization of the complete chloroplast genome sequences of six Dalbergia species and its comparative analysis in the subfamily of Papilionoideae (Fabaceae).

Dalbergia spp. are numerous and widely distributed in pantropical areas in Asia, Africa and America, and most of the species have important economic and ecological value as precious timber. In this study, we determined and characterized six complete chloroplast genomes of Dalbergia species (Dalbergia obtusifolia, D. hupeana, D. mimosoides, D. sissoo, D. hancei, D. balansae), which displayed the typical quadripartite structure of angiosperms. The sizes of the genomes ranged from 155,698 bp (D. hancei) to 156,419 bp (D. obtusifolia). The complete chloroplast genomes of Dalbergia include 37 tRNA genes, eight rRNA genes and 84 protein-coding genes. We analysed the sequence diversity of Dalberigia chloroplast genomes coupled with previous reports. The results showed 12 noncoding regions (rps16-accD, trnR-UCU-trnG-UCC, ndhE-ndhG, trnG-UCC-psbZ, rps8-rpl14, trnP-UGG-psaJ, ndhH-rps15, trnQ-UUG-rps16, trnS-GCU-psbI, rps12-clpP, psbA-trnK-UUU, trnK-UUU-intron), and four coding regions (rps16, ycf1, rps15 and ndhF) showed many nucleotide variations that could be used as potential molecular markers. Based on a site-specific model, we analysed the selective pressure of chloroplast genes in Dalbergia species. Twenty-two genes with positively selected sites were detected, involving the photosynthetic system (ndhC, adhD, ndhF, petB, psaA, psaB, psbB, psbC, psbK and rbcL), self-replication category of genes (rpoA, rpoC2, rps3, rps12 and rps18) and others (accD, ccsA, cemA, clpP, matK, ycf1 and ycf2). Additionally, we identified potential RNA editing sites that were relatively conserved in the genus Dalbergia. Furthermore, the comparative analysis of cp genomes of Dalbergieae species indicated that the boundary of IRs/SSC was highly variable, which resulted in the size variation of cp genomes. Finally, phylogenetic analysis showed an inferred phylogenetic tree of Papilionoideae species with high bootstrap support and suggested that Amorpheae was the sister of the clade Dalbergieae. Moreover, three genera of the Pterocarpus clade showed a nested evolutionary relationship. These complete cp genomes provided valuable information for understanding the genetic variation and phylogenetic relationship of Dalbergia species with their relatives.

Accumulation of Anthocyanidins Determines Leaf Color of Liquidambar Formosana as Revealed by Transcriptome Sequencing and Metabolism Analysis.

Liquidambar formosana is important for its ornamental value in China; it is increasingly used for landscaping and gardening trees due to its diverse leaf colors and seasonal changes. Varieties including either a fixed leaf color, the purplish 'Fuluzifeng' (ZF), or seasonal changes in leaf color, the reddish 'Nanlinhong' (NLH) have been bred and registered as new plant varieties under the International Union for the Protection of New Plant Varieties (UPOV) system. To gain practical insights into the anthocyanin biosynthetic process, transcriptome sequencing (Illumina) was performed to clarify the metabolic pathways present in the three seasonal changes in leaf colors in NLH and in the springtime purple-red color of ZF. qRT-PCR was used to verify the speculation. Based on the differentially expressed genes and flavonoids analyses, the spring, summer, and autumn leaves of NLH were compared to study the seasonal differences. NLH and ZF were compared to study the formation mechanism of the two leaf colors in spring. Transcriptome sequencing produced a total of 121,216 unigenes from all samples, where 48 unigenes were differentially expressed and associated with the anthocyanidin pathway. The expression levels of LfDFR and LfANS genes corresponded to the accumulation of concentrations of cyanidins in spring (NLHC) and autumn leaves (NLHQ), respectively, with different shades of red. Moreover, the LfF3'5'H gene corresponded to the accumulation of flavonols and delphinidins in purple-red leaves (ZFC). Cyanidin and peonidin were the key pigments in red and dark-red leaves, and purple-red leaves were co-pigmented by cyanidins, pelargonidins, and delphinidins.

A high-quality chromosome-level genome of wild Rosa rugosa.

Rosa rugosa is an important shrub with economic, ecological, and pharmaceutical value. A high-quality chromosome-scale genome for R. rugosa sequences was assembled using PacBio and Hi-C technologies. The final assembly genome sequences size was about 407.1 Mb, the contig N50 size was 2.85 Mb, and the scaffold N50 size was 56.6 Mb. More than 98% of the assembled genome sequences were anchored to seven pseudochromosomes (402.9 Mb). The genome contained 37,512 protein-coding genes, with 37,016 genes (98.68%) that were functionally annotated, and 206.67 Mb (50.76%) of the assembled sequences are repetitive sequences. Phylogenetic analyses indicated that R. rugosa diverged from Rosa chinensis ∼6.6 million years ago, and no lineage-specific whole-genome duplication event occurred after divergence from R. chinensis. Chromosome synteny analysis demonstrated highly conserved synteny between R. rugosa and R. chinensis, between R. rugosa and Prunus persica as well. Comparative genome and transcriptome analysis revealed genes related to colour, scent, and environment adaptation. The chromosome-level reference genome provides important genomic resources for molecular-assisted breeding and horticultural comparative genomics research.

Molecular characterization and expression analysis of small heat shock protein 17.3 gene from Sorbus pohuashanensis (Hance) Hedl. in response to abiotic stress.

Sorbus pohuashanensis, a native tree species in China that is distributed at high altitudes. However, the problem of adaptability when introducing S. pohuashanensis to low altitude areas has not been solved. sHSPs can respond and play an essential role when exposing to abiotic stresses for plants. In this study, we aimed to investigate the expression patterns underlying the abiotic stress response of the small heat shock protein 17.3 gene from S. pohuashanensis (SpHSP17.3) at growing low altitude. 1 to 4 years old seedlings of S. pohuashanensis were used as materials for the gene cloning, the tissue-specific expression and the expression analysis underlying the response to abiotic stress using the transgentic methods and qPCR. We identified the open reading frame (ORF) sequence of SpHSP17.3 of 471 bp, which encodes a 17.3 kD protein of 156 amino acids that is located in cytoplasmic. We found that SpHSP17.3 had the highest expression in the stem, followed sequentially by fruit, root, and flower. The expression level of SpHSP17.3 in the leaves was significantly induced by the high temperature (42 °C), NaCl salt and drought stress of S. pohuashanensis. Notably, the same SpHSP17.3 expression trend was detected in the SpHSP17.3-overexpressing homozygous transgenic Arabidopsis underlying the high temperature, NaCl salt and drought stress, and the SpHSP17.3-overexpressing homozygous transgenic Arabidopsis also showed higher seed germination rates under the NaCl salt stress conditions. Our results suggested that SpHSP17.3 is involved in the response to high temperature, Nacl salt, and drought stress which would play a certain effect in the adaptability of introduction and domestication of S. pohuashanensis.