New technology offers tremendous opportunities, but it also presents significant challenges.GSA remains committed to addressing these challenges directly and to publishing work with transparent, rigorous methods. We’re looking for submissions that raise complex questions —whether they present solutions or clearly document biases when solutions are not yet available. 

This month, G3:Genes|Genomes|Genetics features four papers that take on difficult issues and confront tough questions head-on:

Limitations of current machine learning models in predicting enzymatic functions for uncharacterized proteins

There are hundreds of proteins in E.coli whose function is unknown. While the current generation of machine learning models fail to predict beyond the model space, new tools in explainable AI help identify the most common prediction errors. 

 

Estimating allele frequencies, ancestry proportions and genotype likelihoods in the presence of mapping bias

The authors describe a range of practical, readily available and time efficient approaches to identifying and correcting for map-bias.  

Accounting for gene flow from unsampled ghost populations while estimating evolutionary history

Gene flow from unsampled or extinct “ghost” populations leave signatures on the genomes of individuals from extant, sampled populations, often introducing biases, data misinterpretation, and ambiguous results when estimating evolutionary history from population genomic data.

MLDAAPP: machine learning data acquisition for assessing population phenotypes​

An interactive easy-to-use platform for annotating and training image and video data for automatic phenotyping.  

Together, these studies demonstrate how grappling with complexity advances both methods and understanding.

Have a paper that digs into the promises and pitfalls of new technology? We’re looking for work that doesn’t shy away from the tough questions—papers that examine challenges, raise complex issues, and spark thoughtful discussion.

Check out the list of papers below, and if your work grapples with hard questions, we’d love to see it. Submit your papers at https://g3.msubmit.net. We can’t wait to read them!

, , , ,
View all posts by Editorial Staff »

Read more in