Contrast trees and distribution boosting.
Ontology highlight
ABSTRACT:
SUBMITTER: Friedman JH
PROVIDER: S-EPMC7474603 | biostudies-literature | 2020 Sep
REPOSITORIES: biostudies-literature
Publications
Contrast trees and distribution boosting.
Proceedings of the National Academy of Sciences of the United States of America 20200819 35
A method for decision tree induction is presented. Given a set of predictor variables [Formula: see text] and two outcome variables y and z associated with each x, the goal is to identify those values of x for which the respective distributions of [Formula: see text] and [Formula: see text], or selected properties of those distributions such as means or quantiles, are most different. Contrast trees provide a lack-of-fit measure for statistical models of such statistics, or for the complete condi ...[more]
Similar Datasets
Project description:BackgroundThe COVID-19 pandemic has led to an increased demand for health care resources and, in some cases, shortage of medical equipment and staff. Our objective was to develop and validate a multivariable model to predict risk of hospitalization for patients infected with SARS-CoV-2.MethodsWe used routinely collected health records in a patient cohort to develop and validate our prediction model. This cohort included adult patients (age ≥ 18 yr) from Ontario, Canada, who tested positive for SARS-CoV-2 ribonucleic acid by polymerase chain reaction between Feb. 2 and Oct. 5, 2020, and were followed up through Nov. 5, 2020. Patients living in long-term care facilities were excluded, as they were all assumed to be at high risk of hospitalization for COVID-19. Risk of hospitalization within 30 days of diagnosis of SARS-CoV-2 infection was estimated via gradient-boosting decision trees, and variable importance examined via Shapley values. We built a gradient-boosting model using the Extreme Gradient Boosting (XGBoost) algorithm and compared its performance against 4 empirical rules commonly used for risk stratifications based on age and number of comorbidities.ResultsThe cohort included 36 323 patients with 2583 hospitalizations (7.1%). Hospitalized patients had a higher median age (64 yr v. 43 yr), were more likely to be male (56.3% v. 47.3%) and had a higher median number of comorbidities (3, interquartile range [IQR] 2-6 v. 1, IQR 0-3) than nonhospitalized patients. Patients were split into development (n = 29 058, 80.0%) and held-out validation (n = 7265, 20.0%) cohorts. The gradient-boosting model achieved high discrimination (development cohort: area under the receiver operating characteristic curve across the 5 folds of 0.852; validation cohort: 0.8475) and strong calibration (slope = 1.01, intercept = -0.01). The patients who scored at the top 10% captured 47.4% of hospitalizations, and those who scored at the top 30% captured 80.6%.InterpretationWe developed and validated an accurate risk stratification model using routinely collected health administrative data. We envision that modelling such risk stratification based on routinely collected health data could support management of COVID-19 on a population health level.
| S-EPMC8695533 | biostudies-literature
Project description:BackgroundMexico is one of the most floristically rich countries in the world. Despite significant contributions made on the understanding of its unique flora, the knowledge on its diversity, geographic distribution and human uses, is still largely fragmented. Unfortunately, deforestation is heavily impacting this country and native tree species are under threat. The loss of trees has a direct impact on vital ecosystem services, affecting the natural capital of Mexico and people's livelihoods. Given the importance of trees in Mexico for many aspects of human well-being, it is critical to have a more complete understanding of their diversity, distribution, traditional uses and conservation status. We aimed to produce the most comprehensive database and catalogue on native trees of Mexico by filling those gaps, to support their in situ and ex situ conservation, promote their sustainable use, and inform reforestation and livelihoods programmes.MethodsA database with all the tree species reported for Mexico was prepared by compiling information from herbaria and reviewing the available floras. Species names were reconciled and various specialised sources were used to extract additional species information, i.e. endemic status, threat status, availability in seed collections, reports on plant uses and conservation actions currently in place. With this information, a comprehensive catalogue of native trees from Mexico was redacted. Available georeferenced records were used to map each species distribution and perform spatial analyses to identify gaps of information and priority areas for their conservation and exploration.ResultsMexico has at least 2,885 native tree species, belonging to 612 genera and 128 families. Fabaceae is the most represented family and Quercus the most represented genus. Approximately 44% of tree species are endemic to the country. The southern part of the country showed the highest values of species richness. Six hundred and seventy-four species have at least one documented human use. In terms of conservation assessment, ca. 33% of species have been assessed by either the IUCN Red List (919) or the National protection catalogue "NORMA Oficial Mexicana NOM-059" (29) or both (45). Additionally, 98 species have been included in the CITES listing for protection. In terms of existing conservation efforts, 19% of species have ex situ protection in seed banks, while protected areas overlap with all the identified peaks of species richness, except for those in the states of Veracruz and Chiapas. This work constitutes a key milestone for the knowledge, management, and conservation of the Mexican native trees. The two areas with high density of tree species identified in Veracruz and Chiapas represent two priority areas for tree conservation in Mexico, where integrated in situ and ex situ conservation efforts should be focused.
| S-EPMC7505059 | biostudies-literature
Project description:Highlighting patterns of distribution and assembly of plants involves the use of community phylogenetic analyses and complementary traditional taxonomic metrics. However, these patterns are often unknown or in dispute, particularly along elevational gradients, with studies finding different patterns based on elevation. We investigated how patterns of tree diversity and structure change along an elevation gradient using taxonomic and phylogenetic diversity metrics. We sampled 595 individuals (36 families; 53 genera; 88 species) across 15 plots along an elevational gradient (2440-3330 m) in Ecuador. Seventy species were sequenced for the rbcL and matK gene regions to generate a phylogeny. Species richness, Shannon-Weaver diversity, Simpson's Dominance, Simpson's Evenness, phylogenetic diversity (PD), mean pairwise distance (MPD), and mean nearest taxon distance (MNTD) were evaluated for each plot. Values were correlated with elevation and standardized effect sizes (SES) of MPD and MNTD were generated, including and excluding tree fern species, for comparisons across elevation. Taxonomic and phylogenetic metrics found that species diversity decreases with elevation. We also found that overall the community has a non-random phylogenetic structure, dependent on the presence of tree ferns, with stronger phylogenetic clustering at high elevations. Combined, this evidence supports the ideas that tree ferns have converged with angiosperms to occupy the same habitat and that an increased filtering of clades has led to more closely related angiosperm species at higher elevations.
| S-EPMC6783956 | biostudies-literature
Project description:BackgroundLong noncoding RNAs (lncRNAs) play important roles in various biological and pathological processes. Discovery of lncRNA-protein interactions (LPIs) contributes to understand the biological functions and mechanisms of lncRNAs. Although wet experiments find a few interactions between lncRNAs and proteins, experimental techniques are costly and time-consuming. Therefore, computational methods are increasingly exploited to uncover the possible associations. However, existing computational methods have several limitations. First, majority of them were measured based on one simple dataset, which may result in the prediction bias. Second, few of them are applied to identify relevant data for new lncRNAs (or proteins). Finally, they failed to utilize diverse biological information of lncRNAs and proteins.ResultsUnder the feed-forward deep architecture based on gradient boosting decision trees (LPI-deepGBDT), this work focuses on classify unobserved LPIs. First, three human LPI datasets and two plant LPI datasets are arranged. Second, the biological features of lncRNAs and proteins are extracted by Pyfeat and BioProt, respectively. Thirdly, the features are dimensionally reduced and concatenated as a vector to represent an lncRNA-protein pair. Finally, a deep architecture composed of forward mappings and inverse mappings is developed to predict underlying linkages between lncRNAs and proteins. LPI-deepGBDT is compared with five classical LPI prediction models (LPI-BLS, LPI-CatBoost, PLIPCOM, LPI-SKF, and LPI-HNM) under three cross validations on lncRNAs, proteins, lncRNA-protein pairs, respectively. It obtains the best average AUC and AUPR values under the majority of situations, significantly outperforming other five LPI identification methods. That is, AUCs computed by LPI-deepGBDT are 0.8321, 0.6815, and 0.9073, respectively and AUPRs are 0.8095, 0.6771, and 0.8849, respectively. The results demonstrate the powerful classification ability of LPI-deepGBDT. Case study analyses show that there may be interactions between GAS5 and Q15717, RAB30-AS1 and O00425, and LINC-01572 and P35637.ConclusionsIntegrating ensemble learning and hierarchical distributed representations and building a multiple-layered deep architecture, this work improves LPI prediction performance as well as effectively probes interaction data for new lncRNAs/proteins.
| S-EPMC8489074 | biostudies-literature
Project description:The present study with young poplar trees aimed at characterizing the effect of O2 shortage in the soil on net uptake of NO3- and NH4+ and the spatial distribution of the N taken up. Moreover, we assessed biomass increment as well as N status of the trees affected by O2 deficiency. For this purpose, an experiment was conducted in which hydroponically grown young poplar trees were exposed to hypoxic and normoxic (control) conditions for 14 days. 15N-labelled NO3- and NH4+ were used to elucidate N uptake and distribution of currently absorbed N and N allocation rates in the plants. Whereas shoot biomass was not affected by soil O2 deficiency, it significantly reduced root biomass and, consequently, the root-to-shoot ratio. Uptake of NO3- but not of NH4+ by the roots of the trees was severely impaired by hypoxia. As a consequence of reduced N uptake, the N content of all poplar tissues was significantly diminished. Under normoxic control conditions, the spatial distribution of currently absorbed N and N allocation rates differed depending on the N source. Whereas NO3- derived N was mainly transported to the younger parts of the shoot, particularly to the developing and young mature leaves, N derived from NH4+ was preferentially allocated to older parts of the shoot, mainly to wood and bark. Soil O2 deficiency enhanced this differential allocation pattern. From these results we assume that NO3- was assimilated in developing tissues and preferentially used to maintain growth and ensure plant survival under hypoxia, whereas NH4+ based N was used for biosynthesis of storage proteins in bark and wood of the trees. Still, further studies are needed to understand the mechanistic basis as well as the eco-physiological advantages of such differential allocation patterns.
| S-EPMC4550380 | biostudies-literature
Project description:We know surprisingly little about whole-tree nonstructural carbon (NSC; primarily sugars and starch) budgets. Even less well understood is the mixing between recent photosynthetic assimilates (new NSC) and previously stored reserves. And, NSC turnover times are poorly constrained. We characterized the distribution of NSC in the stemwood, branches, and roots of two temperate trees, and we used the continuous label offered by the radiocarbon (carbon-14, (14) C) bomb spike to estimate the mean age of NSC in different tissues. NSC in branches and the outermost stemwood growth rings had the (14) C signature of the current growing season. However, NSC in older aboveground and belowground tissues was enriched in (14) C, indicating that it was produced from older assimilates. Radial patterns of (14) C in stemwood NSC showed strong mixing of NSC across the youngest growth rings, with limited 'mixing in' of younger NSC to older rings. Sugars in the outermost five growth rings, accounting for two-thirds of the stemwood pool, had a mean age < 1 yr, whereas sugars in older growth rings had a mean age > 5 yr. Our results are thus consistent with a previously-hypothesized two-pool ('fast' and 'slow' cycling NSC) model structure. These pools appear to be physically distinct.
| S-EPMC4405048 | biostudies-literature
Project description:Two ecologically and economically important, and threatened Dipterocarp trees Sal (Shorea robusta) and Garjan (Dipterocarpus turbinatus) form mono-specific canopies in dry deciduous, moist deciduous, evergreen, and semievergreen forests across South Asia and continental parts of Southeast Asia. They provide valuable timber and play an important role in the economy of many Asian countries. However, both Dipterocarp trees are threatened by continuing forest clearing, habitat alteration, and global climate change. While climatic regimes in the Asian tropics are changing, research on climate change-driven shifts in the distribution of tropical Asian trees is limited. We applied a bioclimatic modeling approach to these two Dipterocarp trees Sal and Garjan. We used presence-only records for the tree species, five bioclimatic variables, and selected two climatic scenarios (RCP4.5: an optimistic scenario and RCP8.5: a pessimistic scenario) and three global climate models (GCMs) to encompass the full range of variation in the models. We modeled climate space suitability for both species, projected to 2070, using a climate envelope modeling tool "MaxEnt" (the maximum entropy algorithm). Annual precipitation was the key bioclimatic variable in all GCMs for explaining the current and future distributions of Sal and Garjan (Sal: 49.97 ± 1.33; Garjan: 37.63 ± 1.19). Our models predict that suitable climate space for Sal will decline by 24% and 34% (the mean of the three GCMs) by 2070 under RCP4.5 and RCP8.5, respectively. In contrast, the consequences of imminent climate change appear less severe for Garjan, with a decline of 17% and 27% under RCP4.5 and RCP8.5, respectively. The findings of this study can be used to set conservation guidelines for Sal and Garjan by identifying vulnerable habitats in the region. In addition, the natural habitats of Sal and Garjan can be categorized as low to high risk under changing climates where artificial regeneration should be undertaken for forest restoration.
| S-EPMC5383467 | biostudies-literature
Project description:Identifying in advance who is unlikely to respond to a specific antidepressant treatment is crucial to precision medicine efforts. The current work leverages genome-wide genetic variation and machine learning to predict response to the antidepressant citalopram using data from the Sequenced Treatment Alternatives to Relieve Depression (STAR*D) trial (n = 1257 with both valid genomic and outcome data). A confirmatory approach selected 11 SNPs previously reported to predict response to escitalopram in a sample different from the current study. A novel exploratory approach selected SNPs from across the genome using nested cross-validation with elastic net logistic regression with a predominantly lasso penalty (alpha = 0.99). SNPs from each approach were combined with baseline clinical predictors and treatment response outcomes were predicted using a stacked ensemble of gradient boosting decision trees. Using pre-treatment clinical and symptom predictors only, out-of-fold prediction of a novel treatment response definition based on STAR*D treatment guidelines was acceptable, AUC = .659, 95% CI [0.629, 0.689]. The inclusion of SNPs using confirmatory or exploratory selection methods did not improve the out-of-fold prediction of treatment response (AUCs were .662, 95% CI [0.632, 0.692] and .655, 95% CI [0.625, 0.685], respectively). A similar pattern of results were observed for the secondary outcomes of the presence or absence of distressing side effects regardless of treatment response and achieving remission or satisfactory partial response, assuming medication tolerance. In the current study, incorporating SNP variation into prognostic models did not enhance the prediction of citalopram response in the STAR*D sample.
| S-EPMC7881144 | biostudies-literature
Project description:The sequence kernel association test (SKAT) is probably the most popular statistical test used in rare-variant association studies. Its null distribution involves unknown parameters that need to be estimated. The current estimation method has a valid type I error rate, but the power is compromised given that all subjects are used for estimation. I have developed an estimation method that uses only control subjects. Named SKAT+, this method uses the same test statistic as SKAT but differs in the way the null distribution is estimated. Extensive simulation studies and applications to data from the Genetic Analysis Workshop 17 and the Ocular Hypertension Treatment Study demonstrated that SKAT+ has superior power over SKAT while maintaining control over the type I error rate. This method is applicable to extensions of SKAT in the literature.
| S-EPMC5005443 | biostudies-literature
Project description:Venerable trees are important natural resources and cultural heritage, offering historical, ecological, social and economic value. However, global warming and anthropogenic activities have threatened their welfare and survival. A comprehensive understanding of their current and future spatial patterns, vis-á-vis environmental conditions, can inform the co-management of sustainable resource use and conservation. We employed the existing spatial occurrence data and environmental variables (bioclimate and elevation) to simulate the optimal habitats for venerable trees in China’s Sichuan Province. We evaluated the current and future climate scenarios of 2100 with double CO2 concentration. The BIOCLIM and QGIS spatial analyses assessed the primary factors of geographical distribution. The results identified 10,720 venerable trees from 123 species, 81 genera and 42 families. Cupressus funebris dominated, with the maximum importance value, followed by Ginkgo biloba, Ficus virens var. sublanceolata, and Phoebe zhennan. The elevation distribution of tree abundance and species richness demonstrated a unimodal pattern, skewing to the low-elevation end, with a concentration in the 600–1500 m low-medium altitude. The majority of trees and excellent habitats were found in eastern Sichuan with a less harsh terrain and climate. The bio3 (isothermality) and bio7 (temperature annual range) factors significantly influenced tree occurrence. Temperature imposed a greater effect on distribution than moisture under the current climate scenario. For the future climate-change scenario, the suitable habitats were predicted to maintain an overall stable pattern, with largely contiguous expansions of better habitats. However, climate warming would shrink the excellent habitats on the plains. The findings can inform strategies and guidelines for venerable-tree conservation in Sichuan. Furthermore, vulnerable areas could be identified. The future range expansion sites could be enlisted to cultivate new trees to replenish the venerable-tree pool. Habitat patches that remain sustainable could provide refugia with the potential for protected-area designation.
| S-EPMC9780929 | biostudies-literature