Analyses in 2019

The 2018 Human Protein Atlas Image Classification competition sought to improve automated classification of protein subcellular localizations from fluorescence images. The winning strategies involved innovative deep learning approaches for multi-label classification.

One third of verified gene knock outs with CRISPR still show residual protein expression owing to translation reinitiation or exon skipping. Several proteins are still functional. The authors call for a systematic analysis of protein levels after genome editing.

The 2018 Data Science Bowl challenged competitors to develop an accurate tool for segmenting stained nuclei from diverse light microscopy images. The winners deployed innovative deep-learning strategies to realize configuration-free segmentation.

In this DREAM challenge, 75 methods for the identification of disease-relevant modules from molecular networks are compared and validated with GWAS data. The authors provide practical guidelines for users and establish benchmarks for network analysis.

A dataset made up of single cancer cells or their mixtures serves as a benchmark for testing almost 4,000 combinations of scRNA-seq data analysis methods.

Among 17 measures of association tested, measures of proportionality consistently performed well for inference of gene and cellular networks, cell clusters and links to disease from scRNA-seq data. In contrast, several widely used measures of association performed well on only a subset of tasks.

This study reports results from the second community-wide single-molecule localization microscopy software challenge, which tested over 30 software packages on realistic simulated data for multiple popular 3D image acquisition modes, as well as 2D localization microscopy.