LI Detector: a framework for sensitive colony-based screens regardless of the distribution of fitness effects

Abstract Microbial growth characteristics have long been used to investigate fundamental questions of biology. Colony-based high-throughput screens enable parallel fitness estimation thousands individual strains using colony as a proxy for fitness. However, is complicated by spatial biases affecting growth, including uneven nutrient distribution, agar surface irregularities, and batch effects. Analytical methods that developed correct these rely on the following assumptions: (1) effects are normally distributed, (2) most genetic perturbations lead minor changes in Although reasonable many applications, assumptions not always warranted can limit ability detect small Beneficial effects, particular, notoriously difficult under assumptions. Here, we linear interpolation-based detector (LI Detector) framework sensitive colony-based screening without making prior about underlying distribution The LI Detector uses grid reference colonies assign relative value every plate. We show effective correcting equally toward increase decrease offers tunable system allows user identify with unprecedented sensitivity specificity. be utilized develop refine gene–gene gene–environment interaction networks colony-forming organisms, yeast, increasing range reliably detected.