Back

TraDIS-validate: a method for curating ordered gene-replacement libraries

Goodall, E. C. A.; Morris, F. C.; McKeand, S. A.; Sullivan, R.; Warner, I. A.; Sheehan, E.; Boelter, G.; Icke, C.; Cunningham, A. F.; Cole, J. A.; Banzhaf, M.; Bryant, J. A.; Henderson, I. R.

2022-02-16 genetics
10.1101/2022.02.16.479736 bioRxiv
Show abstract

In recent years the availability of genome sequence information has grown logarithmically resulting in the identification of a plethora of uncharacterised genes. To address this gap in functional annotation, many high-throughput screens have been devised with the goal of uncovering novel gene functions. Gene-replacement libraries are one such tool that can be screened in a high-throughput way to link genotype and phenotype and are key community resources. However, for a phenotype to be attributed to a specific gene, there needs to be confidence in the genotype. Construction of large libraries can be laborious and occasionally errors will arise. Here, we present a rapid and accurate method for the validation of any ordered gene-replacement library. We applied our method (TraDIS-Validate) to the well-known Keio library of Escherichia coli gene-deletion mutants. Our method identified 3,718 constructed mutants out of a total of 3,728 confirmed isolates. TraDIS-validate therefore had a success rate of 99.7% for identifying the correct kanamycin cassette position. This dataset provides a benchmark for the purity of the Keio mutants and a screening method for mapping the position of an antibiotic resistance cassette in any ordered library.

Matching journals

The top 2 journals account for 50% of the predicted probability mass.

50% of probability mass above

"Similar papers" are the closest papers from that journal in the model's embedding space. They show what the match is built on, but the ranking comes mostly from a classifier over the whole training set, not from these examples alone.