Pekka Marttinen

OMIDIENT is a multi-omics integration method that uses Dirichlet auto-encoders to improve clustering, classification, and data reconstruction across cancer types.

The paper estimates the basic reproduction number (R0) for pandemic E. coli clones and finds that ST131-A has an R0 comparable to pandemic influenza viruses.

This study evaluates polygenic scoring methods across five biobanks and finds that biobank differences outweigh method differences, with ensemble learning improving performance.

HAPNEST generates and evaluates large-scale synthetic genotype and phenotype datasets for polygenic risk score analysis.

This paper introduces a deep learning method to detect gene-gene interactions by considering all SNPs and complex interactions, outperforming existing methods and identifying novel interactions in a cholesterol study.

This study uses satellite imagery and machine learning to classify crop loss at field parcel scale, achieving an average AUC of 0.688 using random forest.

This paper introduces a regularization method called main effect prior (MEP) to improve genotype-phenotype prediction using GWAS summary statistics, showing improved performance on small datasets.

The paper presents a Bayesian model to estimate bacterial acquisition and clearance rates by accounting for within-host genetic variation and sampling intervals, validated using simulations and applied to MRSA isolates.

The paper shows that using genomic data, machine learning, and historical weather data can predict crop yield in new environments, highlighting non-linear GxE effects and soil-rain interactions.

This study analyzes the genomes of Enterococcus faecium strains to understand their adaptive evolution, identifying plasmids as key indicators of host specificity through population-wide genomic analysis.

Bacmeta simulates genomic evolution in bacterial metapopulations with adjustable connectivity and stochastic events, enabling large-scale simulations and likelihood-free inference.

biMM efficiently estimates genetic variances and covariances for cohorts with high-dimensional phenotype measurements using a bivariate linear mixed model.

The paper introduces fastGEAR, an algorithm for detecting both recent and ancestral recombination in bacterial populations, demonstrating its effectiveness on Streptococcus pneumoniae genomes.

This paper introduces SEER, a statistical method to identify sequence elements associated with bacterial phenotypes, demonstrating its application in discovering antibiotic resistance and invasiveness factors in Streptococcus species.

metaCCA performs multivariate meta-analysis of genome-wide association studies using canonical correlation analysis, showing agreement with pooled individual-level data.

The paper models how host metapopulation structure affects bacterial genetic diversity, showing migration can increase or decrease strain relatedness depending on microepidemic rates.