Variations on a common STRUCTURE: new algorithms for a valuable model

What constitutes a community? A co-occurrence exploration of the Costa Rican avifauna

The concept of a "community" as a form of organization for natural biological systems is both widespread and widely accepted within the ecological and biological sciences. Communities have been defined as groups of organisms that interact in ways that denote interdependence between individuals and taxa (e.g. as defined by "food webs") but they have also been defined as groups of co-occurring organisms that are assumed to interact by virtue of their shared spatiotemporal existence. The latter definition has been debated and challenged in the literature, with mounting evidence for co-occurrence being more indicative of coincident ecological niches in space and time rather than being evidence of ecological interaction or dependency. Using a dataset of 460 Costa Rican bird species divided into breeding and non-breeding season datasets, we empirically demonstrate the ways in which co-occurrence can create illusory communities based on similar occupied ecological niches and similar patterns of co-occurrence at different times of year. We discuss the importance of discerning coincidental co-occurrence from true ecological interactions that would manifest a true community, and further address the importance of differentiating communities of co-occurrence from communities of demonstrable ecological interaction. While co-occurrence is a necessary aspect of interspecific interactions, we discuss and demonstrate here that such co-occurrence does not make a community, nor should explicit patterns of co-occurrence be seen as evidence for evolutionarily important ecological interactions.

Review of the Forensic Applicability of Biostatistical Methods for Inferring Ancestry from Autosomal Genetic Markers

The inference of ancestry has become a part of the services many forensic genetic laboratories provide. Interest in ancestry may be to provide investigative leads or identify the region of origin in cases of unidentified missing persons. There exist many biostatistical methods developed for the study of population structure in the area of population genetics. However, the challenges and questions are slightly different in the context of forensic genetics, where the origin of a specific sample is of interest compared to the understanding of population histories and genealogies. In this paper, the methodologies for modelling population admixture and inferring ancestral populations are reviewed with a focus on their strengths and weaknesses in relation to ancestry inference in the forensic context.

Exploring Population Structure with Admixture Models and Principal Component Analysis

Population structure is a commonplace feature of genetic variation data, and it has importance in numerous application areas, including evolutionary genetics, conservation genetics, and human genetics. Understanding the structure in a sample is necessary before more sophisticated analyses are undertaken. Here we provide a protocol for running principal component analysis (PCA) and admixture proportion inference-two of the most commonly used approaches in describing population structure. Along with hands-on examples with CEPH-Human Genome Diversity Panel and pragmatic caveats, readers will learn to analyze and visualize population structure on their own data.

Profiling SNP and Nucleotide Diversity to Characterize Mekong Delta Rice Landraces in Southeast Asian Populations

Single nucleotide polymorphism (SNP) analyses are a powerful tool to examine structure of local rice population. 3000 dataset of IRRI facilitates SNP profiling of Southeast Asian rice populations. Mekong Delta population is featured by comparisons with the other populations. The low π-value SNPs well-profile unique genetic regions in their genomes. Recent analyses using single nucleotide polymorphism (SNP) are a feasible mean for local collections which potentially possess useful, but not large, genetic variations. Genomic sequences of more than 3000 accessions released by the International Rice Research Institute (IRRI) can be used to characterize various local rice (Oryza sativa) populations. The aim of this study was to develop a method to facilitate genomic characterization of local rice populations. We mainly used 99 indica rice accessions (81 landraces and 18 improved varieties) from the Mekong Delta Development Research Institute (MDI). We obtained 2301 SNPs after a genomic sequencing analysis of the 99 rice accessions and subsequent filtering. Within the IRRI's dataset, the landraces fell into a cluster consisting of accessions from Southeast Asian countries (Ind3 cluster), and the MDI improved varieties were grouped in a cluster containing IRRI improved varieties (Ind1B cluster). A principal component analysis suggested that geographical location strongly affects phylogenetic relationships, and the MDI landraces were placed into a Vietnam+Cambodia group. To detect the nucleotide diversity within a population, π-value is commonly used. We think that whole genome distribution of π-values representing the nucleotide diversity of each population can be used to characterize local populations. Our simple profiling using low π-value genomic regions was able to reveal regional characteristics of rice genomes and should be useful for identifying local rice populations.

A population genetic assessment of taxonomic species: The case of Lake Malawi cichlid fishes

Organisms sampled for population‐level research are typically assigned to species by morphological criteria. However, if those criteria are limited to one sex or life stage, or the organisms come from a complex of closely related forms, the species assignments may misdirect analyses. The impact of such sampling can be assessed from the correspondence of genetic clusters, identified only from patterns of genetic variation, to the species identified using only phenotypic criteria. We undertook this protocol with the rock‐dwelling mbuna cichlids of Lake Malawi, for which species within genera are usually identified using adult male coloration patterns. Given high local endemism of male colour patterns, and considerable allele sharing among species, there persists considerable taxonomic uncertainty in these fishes. Over 700 individuals from a single transect were photographed, genotyped and separately assigned: (a) to morphospecies using photographs; and (b) to genetic clusters using five widely used methods. Overall, the correspondence between clustering methods was strong for larger clusters, but methods varied widely in estimated number of clusters. The correspondence between morphospecies and genetic clusters was also strong for larger clusters, as well as some smaller clusters for some methods. These analyses generally affirm (a) adult male‐limited sampling and (b) the taxonomic status of Lake Malawi mbuna, as the species in our study largely appear to be well‐demarcated genetic entities. More generally, our analyses highlight the challenges for clustering methods when the number of populations is unknown, especially in cases of highly uneven sample sizes.

Reliably discriminating stock structure with genetic markers: Mixture models with robust and fast computation

Delineating naturally occurring and self-sustaining subpopulations (stocks) of a species is an important task, especially for species harvested from the wild. Despite its central importance to natural resource management, analytical methods used to delineate stocks are often, and increasingly, borrowed from superficially similar analytical tasks in human genetics even though models specifically for stock identification have been previously developed. Unfortunately, the analytical tasks in resource management and human genetics are not identical-questions about humans are typically aimed at inferring ancestry (often referred to as "admixture") rather than breeding stocks. In this article, we argue, and show through simulation experiments and an analysis of yellowfin tuna data, that ancestral analysis methods are not always appropriate for stock delineation. In this work, we advocate a variant of a previously introduced and simpler model that identifies stocks directly. We also highlight that the computational aspects of the analysis, irrespective of the model, are difficult. We introduce some alternative computational methods and quantitatively compare these methods to each other and to established methods. We also present a method for quantifying uncertainty in model parameters and in assignment probabilities. In doing so, we demonstrate that point estimates can be misleading. One of the computational strategies presented here, based on an expectation-maximization algorithm with judiciously chosen starting values, is robust and has a modest computational cost.

Fine population structure analysis method for genomes of many

Fine population structure can be examined through the clustering of individuals into subpopulations. The clustering of individuals in large sequence datasets into subpopulations makes the calculation of subpopulation specific allele frequency possible, which may shed light on selection of candidate variants for rare diseases. However, as the magnitude of the data increases, computational burden becomes a challenge in fine population structure analysis. To address this issue, we propose fine population structure analysis (FIPSA), which is an individual-based non-parametric method for dissecting fine population structure. FIPSA maximizes the likelihood ratio of the contingency table of the allele counts multiplied by the group. We demonstrated that its speed and accuracy were superior to existing non-parametric methods when the simulated sample size was up to 5,000 individuals. When applied to real data, the method showed high resolution on the Human Genome Diversity Project (HGDP) East Asian dataset. FIPSA was independently validated on 11,257 human genomes. The group assignment given by FIPSA was 99.1% similar to those assigned based on supervised learning. Thus, FIPSA provides high resolution and is compatible with a real dataset of more than ten thousand individuals.

pong: fast analysis and visualization of latent clusters in population genetic data

MOTIVATION

A series of methods in population genetics use multilocus genotype data to assign individuals membership in latent clusters. These methods belong to a broad class of mixed-membership models, such as latent Dirichlet allocation used to analyze text corpora. Inference from mixed-membership models can produce different output matrices when repeatedly applied to the same inputs, and the number of latent clusters is a parameter that is often varied in the analysis pipeline. For these reasons, quantifying, visualizing, and annotating the output from mixed-membership models are bottlenecks for investigators across multiple disciplines from ecology to text data mining.

RESULTS

We introduce pong, a network-graphical approach for analyzing and visualizing membership in latent clusters with a native interactive D3.js visualization. pong leverages efficient algorithms for solving the Assignment Problem to dramatically reduce runtime while increasing accuracy compared with other methods that process output from mixed-membership models. We apply pong to 225 705 unlinked genome-wide single-nucleotide variants from 2426 unrelated individuals in the 1000 Genomes Project, and identify previously overlooked aspects of global human population structure. We show that pong outpaces current solutions by more than an order of magnitude in runtime while providing a customizable and interactive visualization of population structure that is more accurate than those produced by current tools.

AVAILABILITY AND IMPLEMENTATION

pong is freely available and can be installed using the Python package management system pip. pong's source code is available at https://github.com/abehr/pong

CONTACT

aaron_behr@alumni.brown.edu or sramachandran@brown.edu

SUPPLEMENTARY INFORMATION

Supplementary data are available at Bioinformatics online.