Development and validation of a machine learning model for prediction of type 2 diabetes in patients with mental illness.

BACKGROUND: Type 2 diabetes (T2D) is approximately twice as common among individuals with mental illness compared with the background population, but may be prevented by early intervention on lifestyle, diet, or pharmacologically. Such prevention relies on identification of those at elevated risk (prediction). The aim of this study was to develop and validate a machine learning model for prediction of T2D among patients with mental illness. METHODS: The study was based on routine clinical data from electronic health records from the psychiatric services of the Central Denmark Region. A total of 74,880 patients with 1.59 million psychiatric service contacts were included in the analyses. We created 1343 potential predictors from 51 source variables, covering patient-level information on demographics, diagnoses, pharmacological treatment, and laboratory results. T2D was operationalised as HbA1c ≥48 mmol/mol, fasting plasma glucose ≥7.0 mmol/mol, oral glucose tolerance test ≥11.1 mmol/mol or random plasma glucose ≥11.1 mmol/mol. Two machine learning models (XGBoost and regularised logistic regression) were trained to predict T2D based on 85% of the included contacts. The predictive performance of the best performing model was tested on the remaining 15% of the contacts. RESULTS: The XGBoost model detected patients at high risk 2.7 years before T2D, achieving an area under the receiver operating characteristic curve of 0.84. Of the 996 patients developing T2D in the test set, the model issued at least one positive prediction for 305 (31%). CONCLUSION: A machine learning model can accurately predict development of T2D among patients with mental illness based on routine clinical data from electronic health records. A decision support system based on such a model may inform measures to prevent development of T2D in this high-risk population.

300 Million years of coral treaders (Insecta: Heteroptera: Hermatobatidae) back to the ocean in the phylogenetic context of Arthropoda.

Among hundreds of insect families, Hermatobatidae (commonly known as coral treaders) is one of the most unique. They are small, wingless predaceous bugs in the suborder Heteroptera. Adults are almost black in colour, measuring about 5 mm in body length and 3 mm in width. Thirteen species are known from tropical coral reefs or rocky shores, but their origin and evolutionary adaptation to their unusual marine habitat were unexplored. We report here the genome and metagenome of Hermatobates lingyangjiaoensis, hitherto known only from its type locality in the South China Sea. We further reconstructed the evolutionary history and origin of these marine bugs in the broader context of Arthropoda. The dated phylogeny indicates that Hexapoda diverged from their marine sister groups approximately 498 Ma and that Hermatobatidae originated 192 Ma, indicating that they returned to an oceanic life some 300 Myr after their ancestors became terrestrial. Their origin is consistent with the recovery of tropical reef ecosystems after the end-Triassic mass extinction, which might have provided new and open niches for them to occupy and thrive. Our analyses also revealed that both the genome changes and the symbiotic bacteria might have contributed to adaptations necessary for life in the sea.

Phylogeny and historical biogeography of the water boatmen (Insecta: Hemiptera: Heteroptera: Nepomorpha: Corixoidea).

The water boatmen of Corixoidea, a group of aquatic bugs with more than 600 extant species, is one of the largest superfamilies of Nepomorpha. Contrary to the other nepomorphan lineages, the Corixoidea are most diverse in the Laurasian remnant Holarctic region. To explicitly test whether the present-day Holarctic distribution of diverse corixids is associated with the arising of the Laurasian landmass that was separated from Gondwana, we investigated the phylogeny, divergence times and historical biogeography of Corixoidea based on morphological and molecular characters sampled from 122 taxa representing all families, subfamilies, tribes and approximately 54 % of the genera. Our results were largely congruent with the phylogenetic relationships within the established nepomorphan phylogenetic context. The fossil calibrated chronogram, diversification analysis and ancestral ranges reconstruction indicated that Corixoidea began to diversify in Gondwana in the late Triassic approximately at 224 Ma and the arising of the most diverse subfamily Corixinae in Corixidae in the Holarctic region was largely congruent with the time of separation of Laurasia from Gondwana. The large-scale expansion of the temperate and cold zones on the northward-moving Laurasian landmass after the breakup of the Pangea provided new aquatic niches and ecological opportunities for promoting rapid diversification for the Holarctic corixid lineage.

Climate Warming Since the Holocene Accelerates West-East Communication for the Eurasian Temperate Water Strider Species Aquarius paludum.

Holocene climate warming has dramatically altered biological diversity and distributions. Recent human-induced emissions of greenhouse gases will exacerbate global warming and thus induce threats to cold-adapted taxa. However, the impacts of this major climate change on transcontinental temperate species are still poorly understood. Here, we generated extensive genomic datasets for a water strider, Aquarius paludum, which was sampled across its entire distribution in Eurasia and used these datasets in combination with ecological niche modeling (ENM) to elucidate the influence of the Holocene and future climate warming on its population structure and demographic history. We found that A. paludum consisted of two phylogeographic lineages that diverged in the middle Pleistocene, which resulted in a "west-east component" genetic pattern that was probably triggered by Central Asia-Mongoxin aridification and Pleistocene glaciations. The diverged western and eastern lineages had a second contact in the Holocene, which shaped a temporary hybrid zone located at the boundary of the arid-semiarid regions of China. Future predictions detected a potentially novel northern corridor to connect the western and eastern populations, indicating west-east gene flow would possibly continue to intensify under future warming climate conditions. Further integrating phylogeographic and ENM analyses of multiple Eurasian temperate taxa based on published studies reinforced our findings on the "west-east component" genetic pattern and the predicted future northern corridor for A. paludum. Our study provided a detailed paradigm from a phylogeographic perspective of how transcontinental temperate species differ from cold-adapted taxa in their response to climate warming.

Water bugs (Hemipera: Heteroptera: Nepomorpha amp; Gerromorpha) of Chile: phylogenetic and biogeographic considerations, and a catalog of the fauna.

The Chilean fauna of water bugs comprises seven species of semi-aquatic bugs (Heteroptera: Gerromorpha), representing five genera, three tribes, four subfamilies and four families; and 27 species and one subspecies of aquatic bugs (Heteroptera: Nepomorpha), representing four subgenera, eight genera, three tribes, seven subfamilies, and five families. We compare the fauna with neighboring countries and find that several otherwise widespread and abundant taxa are missing in Chile, but that Chepuvelia usingeri China, 1963 (Macroveliidae), Microvelia chilena Drake Hussey, 1955 (Veliidae), Limnocoris dubiosus Montandon, 1898 (Naucoridae), Nerthra (Nerthra) parvula (Signoret, 1863), N. (N.) undosa Nieser Chen, 1992, N. (Rhinodermacoris) praecipua Todd, 1957 (Gelastocoridae), and Sigara (Tropocorixa) termasensis (Hungerford, 1928a) (Corixidae) are endemic to the country. To this list, we add †Nerthra (Nerthra) subantarctica Faúndez Ashworth, 2015, even though the species is only known from a subfossil. We can also inform that while water bugs are found in the archipelagoes of southern Chile, no species has been reported from the Juan Fernandez Islands, Easter Island and other off-shore islands. Several of the Chilean species are without any close extant relatives, such as C. usingeri and Aquarius chilensis (Berg, 1881) (Gerridae), or with relatives in Oceania (N. praecipua), suggesting that historical events such as dispersal and extinction have had a major influence on the composition of the Chilean fauna.

Phylogeny and diversification of the true water bugs (Insecta: Hemiptera: Heteroptera: Nepomorpha).

Climate fluctuations and tectonic reconfigurations associated with environmental changes play large roles in determining patterns of adaptation and diversification, but studies documenting how such drivers have shaped the evolutionary history and diversification dynamics of limnic organisms during the Mesozoic are scarce. Members of the heteropteran infraorder Nepomorpha, or aquatic bugs, are ideal for testing the effects of these determinants on their diversification pulses because most species are confined to aquatic environments during their entire life. The group has a relatively mature taxonomy and is well represented in the fossil record. We investigated the evolution of Nepomorpha based on phylogenetic analyses of morphological and molecular characters sampled from 115 taxa representing all 13 families and approximately 40% of recognized genera. Our results were largely congruent with the phylogenetic relationships inferred from morphology. A divergence dating analysis indicated that Nepomorpha began to diversify in the late Permian (approximately 263 Ma), and diversification analyses suggested that palaeoecological opportunities probably promoted lineage diversification in this group.

Phylogeny and historical biogeography of Gondwanan moss-bugs (Insecta: Hemiptera: Coleorrhyncha: Peloridiidae).

The moss bugs of the Peloridiidae, a small group of cryptic and mostly flightless insects, is the only living family in Coleorrhyncha (Insecta: Hemiptera). Today 37 species in 17 genera are known from eastern Australia, New Zealand, New Caledonia and Patagonia, and the peloridiids are thereby a group with a classical southern Gondwanan distribution. To explicitly test whether the present-day distribution of the Peloridiidae actually results from the sequential breakup of southern Gondwana, we provide the first total-evidence phylogenetic study based on morphological and molecular characters sampled from about 75% of recognized species representing 13 genera. The results largely confirm the established morphological phylogenetic context except that South American Peloridium hammoniorum constitutes the sister group to the remaining peloridiids. A timescale analysis indicates that the Peloridiidae began to diversify in the land mass that is today's Patagonia in the late Jurassic (153 Ma, 95% highest posterior density: 78-231 Ma), and that splitting into the three extant well-supported biogeographical clades (i.e. Australia, Patagonia and New Zealand/New Caledonia) is consistent with the sequential breakup of southern Gondwana in the late Cretaceous, indicating that the current transoceanic disjunct distributions of the Peloridiidae are best explained by a Gondwanan vicariance hypothesis.

When did the ancestor of true bugs become stinky? Disentangling the phylogenomics of Hemiptera-Heteroptera.

The phylogeny of true bugs (Hemiptera: Heteroptera), one of the most diverse insect groups in terms of morphology and ecology, has been the focus of attention for decades with respect to several deep nodes between the suborders of Hemiptera and the infraorders of Heteroptera. Here, we assembled a phylogenomic data set of 53 taxa and 3102 orthologous genes to investigate the phylogeny of Hemiptera-Heteroptera, and both concatenation and coalescent methods were used. A binode-control approach for data filtering was introduced to reduce the incongruence between different genes, which can improve the performance of phylogenetic reconstruction. Both hypotheses (Coleorrhyncha + Heteroptera) and (Coleorrhyncha + Auchenorrhyncha) received support from various analyses, in which the former is more consistent with the morphological evidence. Based on a divergence time estimation performed on genes with a strong phylogenetic signal, the origin of true bugs was dated to 290-268 Ma in the Permian, the time in Earth's history with the highest concentration of atmospheric oxygen. During this time interval, at least 1007 apomorphic amino acids were retained in the common ancestor of the extant true bugs. These molecular apomorphies are located in 553 orthologous genes, which suggests the common ancestor of the extant true bugs may have experienced large-scale evolution at the genome level.

Water striders (Heteroptera: Gerromorpha: Gerridae) of Romania with an update on the distribution of Gerris gibbifer and G. maculatus in southeastern Europe.

The Romanian fauna comprises two species of Aquarius Schellenberg, 1800, eight species of Gerris Fabricius, 1794 and one species of Limnoporus Stål, 1868, and we hereby update the distribution and provide insights on the phenology and ecology of all eleven species in this country. We furthermore update the distribution of the two closely related species Gerris gibbifer Schummel, 1832 and G. maculatus Tamanini, 1946 in southeastern Europe. Gerris maculatus is recorded for the first time from Hungary, Montenegro and Slovenia, and the first detailed localities from Romania and Serbia are given. All bibliographic records of G. gibbifer from Romania, Macedonia and Serbia are based on misidentification and this species is thus excluded from the faunal lists of these countries. Both G. gibbifer and G. maculatus occur in Croatia, Hungary, Ukraine, and probably Slovenia.

Phylogeography of a semi-aquatic bug, Microvelia horvathi (Hemiptera: Veliidae): an evaluation of historical, geographical and ecological factors.

Subtropical China is a centre of speciation and well known for its high biological diversity and endemism. To understand the impact of historical, geographical and ecological factors on the intraspecific lineage divergence of invertebrates, we examined these processes in a semiaquatic bug, Microvelia horvathi (Hemiptera: Veliidae). Three hypotheses were developed using ecological niche models (ENM). We tested these hypotheses using mitochondrial (COI + COII) and nuclear data (ITS1 + 5.8S + ITS2). The phylogenic analysis revealed a shallow divergence in mitochondrial data. Clade I was mostly confined to the northern region and clade II was nearly restricted to the southern region. The historical process of Pleistocene climatic fluctuations during the LGM promoted divergence, along with such geographical barriers as the Wuyi, Nanling and Xuefeng mountains and ecological factors of temperature and vegetation type, contributed to these shallow genetic divergences and helped maintain them. The north-south population differentiation probably occurred during the transition from LIG to LGM, with post-LGM population expansion. The results of genetic data were mostly consistent with the spatial predictions from ENM. Our study emphasizes the multiple effects influencing genetic population differentiation, and also contributes to our knowledge of the phylogeography of other aquatic organisms in subtropical China.

Phylogenetic divergences of the true bugs (Insecta: Hemiptera: Heteroptera), with emphasis on the aquatic lineages: the last piece of the aquatic insect jigsaw originated in the Late Permian/Early Triassic.

Heteroptera are among the most diverse hemimetabolous insects. Seven infraorders have been recognized within this suborder of Hemiptera. Apart from the well-established sister-group relationship between Cimicomorpha and Pentatomomorpha (= Terheteroptera), the two terminal lineages, the relationships among the other five infraorders are still controversial, of which three (Gerromorpha, Nepomorpha and Leptopodomorpha) are intimately connected to aquatic environments. However, the various and often conflicting available phylogeny hypotheses do not offer a clear background for a connection between diversification and palaeoenvironments. In this study, a molecular data set representing 79 taxa and 10 149 homologous sites is used to infer the phylogenetic relationships within Heteroptera. Bayesian inference, maximum-likelihood and maximum parsimony analyses were employed. The results of phylogenetic inferences largely confirm the widely accepted phylogenetic context. Estimation of the divergence time based on the phylogenetic results revealed that Gerromorpha, Nepomorpha and Leptopodomorpha originated successively during the period from the Late Permian to Early Triassic (269-246 Ma). This timescale is consistent with the origin and radiation time of various aquatic holometabolans. Our results indicate that the aquatic and semi-aquatic true bugs evolved under environmental conditions of high air temperature and humidity in an evolutionary scenario similar to that of the aquatic holometabolans.

A molecular phylogeny of the pan-tropical pond skater genus Limnogonus Stål 1868 (Hemiptera-Heteroptera: Gerromorpha-Gerridae).

We investigated phylogenetic relationships among pond skaters (Heteroptera: Gerridae) of the genus Limnogonus Stål 1868 by performing separate and combined parsimony analyses of DNA sequences from three mitochondrial (COI+II, 16SrRNA) and one nuclear (28SrRNA) gene(s). The taxon sample represented almost two thirds of the known diversity, and with most taxa represented by two or more individuals. A simultaneous analysis of all data showed that L. luctuosus Montrousier 1865 was paraphyletic and suggests that "L. luctuosus" from Australia and possibly also a population from the Society Islands (Moorea) each represents unrecognized species. L. fossarum F. 1775 was strongly supported, but the two subspecies L. f. fossarum F. 1775 and L. f. gilguy Andersen and Weir 1997 were paraphyletic. The two currently recognized subgenera Limnogonus (s. str.) Stål 1868 and L. (Limnogonoides)Andersen 1975 were paraphyletic, and were accordingly broken up in several monophyletic groups, each containing one or more species. From Limnogonus (s. str.) we delimited five clades: I (comprising L. aduncus Drake & Harris 1933, L. recens Drake and Harris 1934, L. profugus Drake & Harris 1930 and L. ignotus Drake and Harris 1934, all from the Neotropical Region), II (comprising L. nitidus Mayr 1865 from the Oriental Region), III (comprising L. franciscanus (Stål 1859) from the New World and L. cereiventris (Signoret 1862) from the Afrotropical Region), IV (comprising L. hungerfordi Andersen 1975 and L. luctuosus Montrousier 1865 from the Oriental and Australasian Regions) and V (comprising L. fossarum F. 1775 from the Oriental and Australasian Regions). From L. (Limnogonoides) we delimited two clades: VI (comprising L. intermedius Poisson 1941 from the Afrotropical Region and L. pectoralis (Mayr 1865) from the Oriental Region) and VII (comprising L. hypoleucus (Gerstaecker 1873), L. nigrescens Poisson 1941, L. poissoni Andersen 1975, and L. capensis China 1925 from the Afrotropical Region). Finally, L. (s. str.) windi Hungerford & Matsuda 1961 from Australia was placed as sister to clades I-VI. A manual optimization of geographical distribution onto the strict consensus tree suggests that Limnogonus is primarily an Old World group with independent transitions to the New World in L. franciscanus and in the common ancestor of Clade I.

Phylogeny of the Heelwalkers (Insecta: Mantophasmatodea) based on mtDNA sequences, with evidence for additional taxa in South Africa.

We examined the phylogeny of Mantophasmatodea from southern Africa (South Africa, Namibia) using approx. 1300 bp of mitochondrial DNA sequence data from the genes encoding COI and 16S. The taxon sample comprised multiple specimens from eight described species (Namaquaphasma ookiepense, Austrophasma rawsonvillense, A. caledonense, A. gansbaaiense, Lobatophasma redelinghuysense, Hemilobophasma montaguense, Karoophasma botterkloofense, K. biedouwense) and four undescribed species of Austrophasmatidae; three specimens of Sclerophasma paresisense (Mantophasmatidae); and two specimens of Praedatophasma maraisi and one of Tyrannophasma gladiator (not yet convincingly assigned to any family). For outgroup comparison a broad selection from hemimetabolous insect orders was included. Equally weighted parsimony analyses of the combined COI+16S data sets with gaps in 16S scored as a fifth character state supported Austrophasmatidae and all species and genera of Mantophasmatodea as being monophyletic. Most species were highly supported with 98-100% bootstrap/7-39 Bremer support (BS), but K. biedouwense had moderate support (87/4) and A. caledonense low support (70/1). Mantophasmatodea, Austrophasmatidae, and a clade Tyrannophasma gladiator+Praedatophasma maraisi were all strongly supported (99-100/12-25), while relationships among the two latter clades and Mantophasmatidae remain ambiguous. Concerning the relationships among genera of Austrophasmatidae, support values are moderately high for some nodes, but not significant for others. We additionally calculated the partitioned BS values of COI and 16S for all nodes in the strict consensus of the combined tree. COI and 16S are highly congruent at the species level as well as at the base of Mantophasmatodea, but congruence is poor for most intergeneric relationships. In forthcoming studies, deeper relationships in the order should be additionally explored by nuclear genes, such as 18S and 28S, for a reduced sample of specimens.

Phylogeny of Mysis (Crustacea, Mysida): history of continental invasions inferred from molecular and morphological data.

We studied the phylogenetic history of opossum shrimps of the genus Mysis Latreille, 1802 (Crustacea: Mysida) using parsimony analyses of morphological characters, DNA sequence data from mitochondrial (16S, COI and CytB) and nuclear genes (ITS2, 18S), and eight allozyme loci. With these data we aimed to resolve a long-debated question of the origin of the non-marine (continental) taxa in the genus, i.e., "glacial relicts" in circumpolar postglacial lakes and "arctic immigrants" in the Caspian Sea. A simultaneous analysis of the data sets gave a single tree supporting monophyly of all continental species, as well as monophyly of the taxa from circumpolar lakes and from the Caspian Sea. A clade of three circumarctic marine species was sister group to the continental taxa, whereas Atlantic species had more distant relationships to the others. Small molecular differentiation among the morphologically diverse endemic species from the Caspian Sea suggested their recent speciation, while the phenotypically more uniform "glacial relict" species from circumpolar lakes (Mysis relicta group) showed deep molecular divergences. For the length-variable ITS2 region both direct optimization and a priori alignment procedures gave similar topologies, although the former approach provided a better overall resolution. In terms of partitioned Bremer support (PBS), mitochondrial protein coding genes provided the largest contribution (83%) to the total tree resolution. This estimate however, appears to be partly spurious, due to the concerted inheritance of mitochondrial characters and probable cases of introgression or ancestral polymorphism.

Sources of character conflict in a clade of water striders (Heteroptera: Gerridae).

Incongruence among trees reconstructed with different data may stem from historical (gene tree-species tree conflict) or process (character change biases) phenomena. Regardless of the source, incongruent data, as determined with "global" measures of homoplasy, have often been excluded from parsimony analysis of the combined data. Recent studies suggest that these homoplasy measures do not predict the contribution of each character to overall tree structure. Branch support measures identify, on a character to node basis, sources of support and conflict resulting from a simultaneous analysis of the data. We implement these branch support measures to identify sources of character conflict in a clade of water striders consisting of Gerris Fabricius, Aquarius Schellenberg, and Limnoporus Stål species. Separate analyses of morphology, mitochondrial cytochrome oxidase I (COI), large mitochondrial ribosomal subunit (16SrRNA), and elongation factor-1α (EF-1α) data resulted in cladograms that varied in resolution and topological concordance. Simultaneous analysis of the data resulted in two trees that were unresolved for one node in a strict consensus. The topology agreed with current classification except for the placements of Aquarius chilensis and the Aquarius remigis species group closer to Gerris than to congeneric species. Branch support measures indicated that support derived from each data set varied among nodes, but COI had an overall negative effect on branch support. However, Spearman rank correlation of partitioned branch support values indicated no negative associations of branch support between any data sets and a positive association between EF-1α and 16SrRNA. Thus incongruence among data sets was not drastic and the gene-tree versus species tree phenomenon was not implicated. Biases in character change were a more likely reason for incongruence, although saturation curves and incongruence length difference for COI indicated little potential for homoplasy. However, a posteriori inspection of COI nucleotide change with reference to the simultaneous analysis tree revealed AT and codon biases. These biases were not associated with branch support measures. Therefore, it is difficult to predict incongruence or identify its cause. Exclusion of data is ill advised because every character is potentially parsimony informative.