Beware the Jaccard: the choice of similarity measure is important and non-trivial in genomic colocalisation analysis

2019 ◽  
Vol 21 (5) ◽  
pp. 1523-1530 ◽  
Author(s):  
Stefania Salvatore ◽  
Knut Dagestad Rand ◽  
Ivar Grytten ◽  
Egil Ferkingstad ◽  
Diana Domanska ◽  
...  
Keyword(s):  
Similarity Measure ◽  
Similarity Measures ◽  
Real Data ◽  
Phenotypic Traits ◽  
Wide Binary ◽  
Genome Wide ◽  
Dataset Size ◽  
Genome Wide Data ◽  
Modelling Assumptions ◽  
Systematic Collection

Abstract The generation and systematic collection of genome-wide data is ever-increasing. This vast amount of data has enabled researchers to study relations between a variety of genomic and epigenomic features, including genetic variation, gene regulation and phenotypic traits. Such relations are typically investigated by comparatively assessing genomic co-occurrence. Technically, this corresponds to assessing the similarity of pairs of genome-wide binary vectors. A variety of similarity measures have been proposed for this problem in other fields like ecology. However, while several of these measures have been employed for assessing genomic co-occurrence, their appropriateness for the genomic setting has never been investigated. We show that the choice of similarity measure may strongly influence results and propose two alternative modelling assumptions that can be used to guide this choice. On both simulated and real genomic data, the Jaccard index is strongly altered by dataset size and should be used with caution. The Forbes coefficient (fold change) and tetrachoric correlation are less influenced by dataset size, but one should be aware of increased variance for small datasets. All results on simulated and real data can be inspected and reproduced at https://hyperbrowser.uio.no/sim-measure.

scholarly journals Beware the Jaccard: the choice of metric is important and non-trivial in genomic colocalisation analysis

2018 ◽  
Author(s):  
Stefania Salvatore ◽  
Knut Dagestad Rand ◽  
Ivar Grytten ◽  
Egil Ferkingstad ◽  
Diana Domanska ◽  
...  
Keyword(s):  
Genomic Data ◽  
Real Data ◽  
Jaccard Index ◽  
Phenotypic Traits ◽  
Wide Binary ◽  
Genome Wide ◽  
Dataset Size ◽  
Genome Wide Data ◽  
Modelling Assumptions ◽  
Systematic Collection

AbstractBackgroundThe generation and systematic collection of genome-wide data is ever-increasing. This vast amount of data has enabled researchers to study relations between a variety of genomic and epigenomic features, including genetic variation, gene regulation, and phenotypic traits. Such relations are typically investigated by comparatively assessing genomic co-occurrence. Technically, this corresponds to assessing the similarity of pairs of genome-wide binary vectors. A variety of metrics have been proposed for this problem in other fields like ecology. However, while several of these metrics have been employed for assessing genomic co-occurrence, their appropriateness for the genomic setting has never been investigated.ResultsWe show that the choice of metric may strongly influence results and propose two alternative modelling assumptions that can be used to guide this choice. On both simulated and real genomic data, the Jaccard index is strongly affected by dataset size and should be used with caution. The Forbes coefficient (fold change) and tetrachoric correlation are less affected by dataset size, but one should be aware of increased variance for small datasets.AvailabilityAll results on simulated and real data can be inspected and reproduced athttps://hyperbrowser.uio.no/sim-measure

scholarly journals A performance assessment of relatedness inference methods using genome-wide data from thousands of relatives

2017 ◽  
Author(s):  
Monica D. Ramstetter ◽  
Thomas D. Dyer ◽  
Donna M. Lehman ◽  
Joanne E. Curran ◽  
Ravindranath Duggirala ◽  
...  
Keyword(s):  
State Of The Art ◽  
Association Studies ◽  
Genetic Association Studies ◽  
Real Data ◽  
New Methods ◽  
Genome Wide ◽  
Genome Wide Data ◽  
Inference Methods ◽  
Multiple Samples ◽  
Combining Information

AbstractInferring relatedness from genomic data is an essential component of genetic association studies, population genetics, forensics, and genealogy. While numerous methods exist for inferring relatedness, thorough evaluation of these approaches in real data has been lacking. Here, we report an assessment of 12 state-of-the-art pairwise relatedness inference methods using a dataset with 2,485 individuals contained in several large pedigrees that span up to six generations. We find that all methods have high accuracy (~92% – 99%) when detecting first and second degree relationships, but their accuracy dwindles to less than 43% for seventh degree relationships. However, most IBD segment-based methods inferred seventh degree relatives correct to within one relatedness degree for more than 76% of relative pairs. Overall, the most accurate methods are ERSA and approaches that compute total IBD sharing using the output from GERMLINE and Refined IBD to infer relatedness. Combining information from the most accurate methods provides little accuracy improvement, indicating that novel approaches—such as new methods that leverage relatedness signals from multiple samples—are needed to achieve a sizeable jump in performance.

scholarly journals ClusTrack: Feature Extraction and Similarity Measures for Clustering of Genome-Wide Data Sets

PLoS ONE ◽  
2015 ◽  
Vol 10 (4) ◽  
pp. e0123261 ◽  
Author(s):  
Halfdan Rydbeck ◽  
Geir Kjetil Sandve ◽  
Egil Ferkingstad ◽  
Boris Simovski ◽  
Morten Rye ◽  
...  
Keyword(s):  
Feature Extraction ◽  
Similarity Measures ◽  
Data Sets ◽  
Genome Wide ◽  
Genome Wide Data

scholarly journals M3C: Monte Carlo reference-based consensus clustering

2018 ◽  
Author(s):  
Christopher R. John ◽  
David Watson ◽  
Dominic Russ ◽  
Katriona Goldmann ◽  
Michael Ehrenstein ◽  
...  
Keyword(s):  
Monte Carlo ◽  
Clustering Algorithm ◽  
Clustering Algorithms ◽  
Real Data ◽  
The Cancer Genome Atlas ◽  
Consensus Clustering ◽  
Null Distributions ◽  
Genome Wide ◽  
Genome Wide Data ◽  
Cancer Genome Atlas

AbstractGenome-wide data is used to stratify patients into classes for precision medicine using clustering algorithms. A common problem in this area is selection of the number of clusters (K). The Monti consensus clustering algorithm is a widely used method which uses stability selection to estimate K. However, the method has bias towards higher values of K and yields high numbers of false positives. As a solution, we developed Monte Carlo reference-based consensus clustering (M3C), which is based on this algorithm. M3C simulates null distributions of stability scores for a range of K values thus enabling a comparison with real data to remove bias and statistically test for the presence of structure. M3C corrects the inherent bias of consensus clustering as demonstrated on simulated and real expression data from The Cancer Genome Atlas (TCGA). For testing M3C, we developed clusterlab, a new method for simulating multivariate Gaussian clusters.

scholarly journals Triplet-based similarity score for fully multi-labeled trees with poly-occurring labels

2020 ◽  
Author(s):  
Simone Ciccolella ◽  
Giulia Bernardini ◽  
Luca Denti ◽  
Paola Bonizzoni ◽  
Marco Previtali ◽  
...  
Keyword(s):  
Similarity Measure ◽  
Evolutionary History ◽  
Similarity Measures ◽  
Real Data ◽  
Similarity Score ◽  
Labeled Trees ◽  
Wide Range ◽  
Golden Standard ◽  
History Of

AbstractThe latest advances in cancer sequencing, and the availability of a wide range of methods to infer the evolutionary history of tumors, have made it important to evaluate, reconcile and cluster different tumor phylogenies.Recently, several notions of distance or similarities have been proposed in the literature, but none of them has emerged as the golden standard. Moreover, none of the known similarity measures is able to manage mutations occurring multiple times in the tree, a circumstance often occurring in real cases.To overcome these limitations, in this paper we propose MP3, the first similarity measure for tumor phylogenies able to effectively manage cases where multiple mutations can occur at the same time and mutations can occur multiple times. Moreover, a comparison of MP3 with other measures shows that it is able to classify correctly similar and dissimilar trees, both on simulated and on real data.

scholarly journals ONLINE FUZZY CLUSTERING OF INCOMPLETE DATA USING CREDIBILISTIC APPROACH AND SIMILARITY MEASURE OF SPECIAL TYPE

2021 ◽  
Vol 1 (1) ◽  
pp. 97-104
Author(s):  
Ye. V. Bodyanskiy ◽  
A. Yu. Shafronenko ◽  
I. N. Klymova
Keyword(s):  
Similarity Measure ◽  
Fuzzy Clustering ◽  
Incomplete Data ◽  
Similarity Measures ◽  
Real Data ◽  
Credibility Theory ◽  
Clustering And Classification ◽  
Clustering Data ◽  
Goal Functions ◽  
Distorted Data

Context. In most clustering (classification without a teacher) tasks associated with real data processing, the initial information is usually distorted by abnormal outliers (noise) and gaps. It is clear that “classical” methods of artificial intelligence (both batch and online) are ineffective in this situation.The goal of the paper is to propose the procedure of fuzzy clustering of incomplete data using credibilistic approach and similarity measure of special type. Objective. The goal of the work is credibilistic fuzzy clustering of distorted data, using of credibility theory. Method. The procedure of fuzzy clustering of incomplete data using credibilistic approach and similarity measure of special type based on the use of both robust goal functions of a special type and similarity measures, insensitive to outliers and designed to work both in batch and its recurrent online version designed to solve Data Stream Mining problems when data are fed to processing sequentially in real time. Results. The introduced methods are simple in numerical implementation and are free from the drawbacks inherent in traditional methods of probabilistic and possibilistic fuzzy clustering data distorted by abnormal outliers (noise) and gaps. Conclusions. The conducted experiments have confirmed the effectiveness of proposed methods of credibilistic fuzzy clustering of distorted data operability and allow recommending it for use in practice for solving the problems of automatic clusterization of distorted data. The proposed method is intended for use in hybrid systems of computational intelligence and, above all, in the problems of learning artificial neural networks, neuro-fuzzy systems, as well as in the problems of clustering and classification.

MATHURA (MBI) - A NOVEL IMPUTATION MEASURE FOR IMPUTATION OF MISSING VALUES IN MEDICAL DATASETS

2019 ◽  
Vol 13 ◽  
Author(s):  
B. Mathura Bai ◽  
N. Mangathayaru ◽  
B. Padmaja Rani ◽  
Shadi Aljawarneh
Keyword(s):  
Similarity Measure ◽  
Medical Records ◽  
Missing Values ◽  
Similarity Measures ◽  
Common Problems ◽  
Experiment Analysis

: Missing attribute values in medical datasets are one of the most common problems faced when mining medical datasets. Estimation of missing values is a major challenging task in pre-processing of datasets. Any wrong estimate of missing attribute values can lead to inefficient and improper classification thus resulting in lower classifier accuracies. Similarity measures play a key role during the imputation process. The use of an appropriate and better similarity measure can help to achieve better imputation and improved classification accuracies. This paper proposes a novel imputation measure for finding similarity between missing and non-missing instances in medical datasets. Experiments are carried by applying both the proposed imputation technique and popular benchmark existing imputation techniques. Classification is carried using KNN, J48, SMO and RBFN classifiers. Experiment analysis proved that after imputation of medical records using proposed imputation technique, the resulting classification accuracies reported by the classifiers KNN, J48 and SMO have improved when compared to other existing benchmark imputation techniques.

scholarly journals Genetic analysis of amyotrophic lateral sclerosis identifies contributing pathways and cell types

2021 ◽  
Vol 7 (3) ◽  
pp. eabd9036
Author(s):  
Sara Saez-Atienzar ◽  
Sara Bandres-Ciga ◽  
Rebekah G. Langston ◽  
Jonggeol J. Kim ◽  
Shing Wan Choi ◽  
...  
Keyword(s):  
Amyotrophic Lateral Sclerosis ◽  
Membrane Trafficking ◽  
Molecular Mechanisms ◽  
Cell Types ◽  
Polygenic Risk Score ◽  
Genome Wide ◽  
Genome Wide Data ◽  
Data Driven Approach ◽  
Single Nucleus ◽  
Lateral Sclerosis

Despite the considerable progress in unraveling the genetic causes of amyotrophic lateral sclerosis (ALS), we do not fully understand the molecular mechanisms underlying the disease. We analyzed genome-wide data involving 78,500 individuals using a polygenic risk score approach to identify the biological pathways and cell types involved in ALS. This data-driven approach identified multiple aspects of the biology underlying the disease that resolved into broader themes, namely, neuron projection morphogenesis, membrane trafficking, and signal transduction mediated by ribonucleotides. We also found that genomic risk in ALS maps consistently to GABAergic interneurons and oligodendrocytes, as confirmed in human single-nucleus RNA-seq data. Using two-sample Mendelian randomization, we nominated six differentially expressed genes (ATG16L2, ACSL5, MAP1LC3A, MAPKAPK3, PLXNB2, and SCFD1) within the significant pathways as relevant to ALS. We conclude that the disparate genetic etiologies of this fatal neurological disease converge on a smaller number of final common pathways and cell types.

scholarly journals Ancient genomic time transect from the Central Asian Steppe unravels the history of the Scythians

2021 ◽  
Vol 7 (13) ◽  
pp. eabe4414
Author(s):  
Guido Alberto Gnecchi-Ruscone ◽  
Elmira Khussainova ◽  
Nurzhibek Kahbatkyzy ◽  
Lyazzat Musralina ◽  
Maria A. Spyrou ◽  
...  
Keyword(s):  
Bronze Age ◽  
Iron Age ◽  
Gene Pools ◽  
Social Rules ◽  
Eurasian Steppe ◽  
Central Asian ◽  
Genome Wide ◽  
Genome Wide Data ◽  
History Of ◽  
First Millennium

The Scythians were a multitude of horse-warrior nomad cultures dwelling in the Eurasian steppe during the first millennium BCE. Because of the lack of first-hand written records, little is known about the origins and relations among the different cultures. To address these questions, we produced genome-wide data for 111 ancient individuals retrieved from 39 archaeological sites from the first millennia BCE and CE across the Central Asian Steppe. We uncovered major admixture events in the Late Bronze Age forming the genetic substratum for two main Iron Age gene-pools emerging around the Altai and the Urals respectively. Their demise was mirrored by new genetic turnovers, linked to the spread of the eastern nomad empires in the first centuries CE. Compared to the high genetic heterogeneity of the past, the homogenization of the present-day Kazakhs gene pool is notable, likely a result of 400 years of strict exogamous social rules.

scholarly journals Genome diversity in Ukraine

GigaScience ◽  
2021 ◽  
Vol 10 (1) ◽  
Author(s):  
Taras K Oleksyk ◽  
Walter W Wolfsberger ◽  
Alexandra M Weber ◽  
Khrystyna Shchubelka ◽  
Olga T Oleksyk ◽  
...  
Keyword(s):  
Sequence Data ◽  
Copy Number Variations ◽  
Genomic Variation ◽  
High Coverage ◽  
Genome Data ◽  
New Information ◽  
Genome Wide ◽  
Public Data ◽  
Genome Wide Data ◽  
Multiple Samples

Abstract Background The main goal of this collaborative effort is to provide genome-wide data for the previously underrepresented population in Eastern Europe, and to provide cross-validation of the data from genome sequences and genotypes of the same individuals acquired by different technologies. We collected 97 genome-grade DNA samples from consented individuals representing major regions of Ukraine that were consented for public data release. BGISEQ-500 sequence data and genotypes by an Illumina GWAS chip were cross-validated on multiple samples and additionally referenced to 1 sample that has been resequenced by Illumina NovaSeq6000 S4 at high coverage. Results The genome data have been searched for genomic variation represented in this population, and a number of variants have been reported: large structural variants, indels, copy number variations, single-nucletide polymorphisms, and microsatellites. To our knowledge, this study provides the largest to-date survey of genetic variation in Ukraine, creating a public reference resource aiming to provide data for medical research in a large understudied population. Conclusions Our results indicate that the genetic diversity of the Ukrainian population is uniquely shaped by evolutionary and demographic forces and cannot be ignored in future genetic and biomedical studies. These data will contribute a wealth of new information bringing forth a wealth of novel, endemic and medically related alleles.

Sign in / Sign up

Export Citation Format

Share Document