WorMachine: machine learning-based phenotypic analysis tool for worms

Abstract

BackgroundCaenorhabditis elegans nematodes are powerful model organisms, yet quantification of visible phenotypes is still often labor-intensive, biased, and error-prone. We developed WorMachine, a three-step MATLAB-based image analysis software that allows (1) automated identification of C. elegans worms, (2) extraction of morphological features and quantification of fluorescent signals, and (3) machine learning techniques for high-level analysis.ResultsWe examined the power of WorMachine using five separate representative assays: supervised classification of binary-sex phenotype, scoring continuous-sexual phenotypes, quantifying the effects of two different RNA interference treatments, and measuring intracellular protein aggregation.ConclusionsWorMachine is suitable for analysis of a variety of biological questions and provides an accurate and reproducible analysis tool for measuring diverse phenotypes. It serves as a “quick and easy,” convenient, high-throughput, and automated solution for nematode research.Electronic supplementary materialThe online version of this article (doi:10.1186/s12915-017-0477-0) contains supplementary material, which is available to authorized users.

Three Rules Explain Transgenerational Small RNA Inheritance in C. elegans

“…The progeny of non-silencer worms show pronounced variability and developmental delay following recovery from 8 days of starvation. Left panel: length measurements ( Hakim et al., 2018 ) 60 h after transferring the worms to plates with food. Differences in variance were assessed by F test.…”

“…After 60 hours of recovery, each group of worms was washed, paralyzed, pictured and then analyzed using the WorMachine as previously described ( Hakim et al., 2018 ), using the WorMachine software: Worms were washed three times to get eliminate bacterial (OP50) residues. Worms were left in ∼100 μL of M9 buffer and paralyzed via the addition of sodium azide (final concentration of 25–50 mM).…”

Deep learning-enabled analysis reveals distinct neuronal phenotypes induced by aging and cold-shock

“…San-Miguel et al implemented a deep phenotyping pipeline to study synaptic patterning in the DA9 motoneuron [ 43 ]. Hakim et al developed a platform called WorMachine which is comprised of image processing, deep learning, and machine learning techniques to perform assays such as supervised classification of binary-sex phenotype, scoring continuous-sexual phenotypes, quantifying the effects of two different RNA interference treatments, and measuring intracellular protein aggregation [ 46 ]. Kaltdorf et al combined a machine learning technique with an image segmentation workflow to develop an automated method to classify Clear Core (CCV) and Dense Core (DCV) synaptic vesicles [ 47 ].…”

Three Rules Explain Transgenerational Small RNA Inheritance inC. elegans

“…Measuring worms’ size using WorMachine: After 60 hours of recovery, each group of worms was washed, paralyzed, pictured and then analyzed as previously described[45], using the WorMachine software.…”

“…Progeny of non-silencer worms show pronounced variability and developmental delay following recovery from 8 days of starvation. Top panel: length measurements [45] 60 hours after transferring the worms to plates with food (F test). Bottom panel: frequency of developmentally delayed worms (arrest as L1/L2) 48 hours after transferring the worms to plates with food (χ 2 test).…”