Enhancing estimation methods for integrating probability and nonprobability survey samples with machine-learning techniques. An application to a Survey on the impact of the COVID-19 pandemic in Spain

Web surveys have replaced Face-to-Face and computer assisted telephone interviewing (CATI) as the main mode of data collection in most countries. This trend was reinforced as a consequence of COVID-19 pandemic-related restrictions. However, this mode still faces significant limitations in obtaining probability-based samples of the general population. For this reason, most web surveys rely on nonprobability survey designs. Whereas probability-based designs continue to be the gold standard in survey sampling, nonprobability web surveys may still prove useful in some situations. For instance, when small subpopulations are the group under study and probability sampling is unlikely to meet sample size requirements, complementing a small probability sample with a larger nonprobability one may improve the efficiency of the estimates. Nonprobability samples may also be designed as a mean for compensating for known biases in probability-based web survey samples by purposely targeting respondent profiles that tend to be underrepresented in these surveys. This is the case in the Survey on the impact of the COVID-19 pandemic in Spain (ESPACOV) that motivates this paper. In this paper, we propose a methodology for combining probability and nonprobability web-based survey samples with the help of machine-learning techniques. We then assess the efficiency of the resulting estimates by comparing them with other strategies that have been used before. Our simulation study and the application of the proposed estimation method to the second wave of the ESPACOV Survey allow us to conclude that this is the best option for reducing the biases observed in our data.     

RUNAWAY COEVOLUTION: ADAPTATION TO HERITABLE AND NONHERITABLE ENVIRONMENTS

Populations evolve in response to the external environment, whether abiotic (e.g., climate) or biotic (e.g., other conspecifics). We investigated how adaptation to biotic, heritable environments differs from adaptation to abiotic, nonheritable environments. We found that, for the same selection coefficients, the coadaptive process between genes and heritable environments is much faster than genetic adaptation to an abiotic nonheritable environment. The increased rate of adaptation results from the positive association generated by reciprocal selection between the heritable environment and the genes responding to it. These associations result in a runaway process of adaptive coevolution, even when the genes creating the heritable environment and genes responding to the heritable environment are unlinked. Although tightening the degree of linkage accelerates the coadaptive process, the acceleration caused by a comparable amount of inbreeding is greater, because inbreeding has a cumulative effect on reducing functional recombination over generations. Our results suggest that that adaptation to local abiotic environmental variation may result in the rapid diversification of populations and subsequent reproductive isolation not directly but rather via its effects on heritable environments and the genes responding to them.     

Variation in the life history pattern of Tetranychus urticae (Acari: Tetranychidae) after selection for dispersal

The spider mite Tetranychus urticae shows variation in its dispersal capacity (i.e., the leaf quality at which a female decides to disperse). We were able to artificially select mites that had either a high or a low dispersal capacity, indicating that this trait was genetically controlled. We then compared correlated responses to this selection. Mites with a genetically high dispersal capacity (‘HD’ strains) had a higher diapause incidence and a lower performance compared to mites with a low dispersal capacity (‘LD’ strains). A possible effect of random genetic drift during the selection was negligible. Our results suggest that differential dispersal capacity is associated with contrasting life history patterns as a result of natural selection. This revised version was published online in July 2006 with corrections to the Cover Date.     

Genetic relatedness in groups is sex-specific and declines with age of helpers in a cooperatively breeding cichlid

Kin selection can explain the evolution of cooperative breeding and the distribution of relatives within a population may influence the benefits of cooperative behaviour. We provide genetic data on relatedness in the cooperatively breeding cichlid Neolamprologus pulcher. Helper to breeder relatedness decreased steeply with increasing helper age, particularly to the breeding males. Helper to helper relatedness was age‐assortative and also declined with age. These patterns of relatedness could be attributed to territory take‐overs by outsiders when breeders had disappeared (more in breeding males), between‐group dispersal of helpers and reproductive parasitism. In six of 31 groups females inherited the breeding position of their mother or sister. These matrilines were more likely to occur in large groups. We conclude that the relative fitness benefits of helping gained through kin selection vs. those gained through direct selection depend on helper age and sex.     

Biological implications of genetic variations in autism spectrum disorders from genomics studies

Autism spectrum disorder (ASD) is a highly heterogeneous neurodevelopmental condition characterized by atypical social interaction and communication together with repetitive behaviors and restricted interests. The prevalence of ASD has been increased these years. Compelling evidence has shown that genetic factors contribute largely to the development of ASD. However, knowledge about its genetic etiology and pathogenesis is limited. Broad applications of genomics studies have revealed the importance of gene mutations at protein-coding regions as well as the interrupted non-coding regions in the development of ASD. In this review, we summarize the current evidence for the known molecular genetic basis and possible pathological mechanisms as well as the risk genes and loci of ASD. Functional studies for the underlying mechanisms are also implicated. The understanding of the genetics and genomics of ASD is important for the genetic diagnosis and intervention for this condition.     

Promiscuous females avoid inbreeding by controlling sperm storage

Recent studies in a variety of species have shown that polyandrous females are somehow able to bias paternity against their relatives postcopulation, although how they do so remains unknown. Field crickets readily mate with their siblings, but when also mated to an unrelated male, they produce disproportionately fewer inbred offspring. We use a new competitive microsatellite polymerase chain reaction technique to determine the contribution of males to stored sperm and subsequent paternity of offspring. Paternity is almost completely predicted by how much sperm from a particular male is stored, and unrelated males contribute more sperm to storage and have a corresponding higher paternity success.     

Isolation and characterization of 23 polymorphic microsatellite loci for a West Indian iguana (Cyclura pinguis) from the British Virgin Islands

Twenty-three polymorphic microsatellite markers were identified and characterized for Cyclura pinguis, a critically endangered species of lizard (Sauria: Iguanidae) native to Anegada Island in the British Virgin Islands. We examined variation at these loci for 39 C. pinguis, finding up to five alleles per locus and an average expected heterozygosity of 0.55. Allele frequency estimates for these microsatellite loci will be used to characterize genetic diversity of captive and wild C. pinguis populations and to estimate relatedness among adult iguanas at the San Diego Zoo that form the nucleus of a captive breeding programme for this critically endangered species.