LaneRuler: Automated Lane Tracking for DNA Electrophoresis Gel Images

We present a novel method for correctly identifying and straightening one dimensional agarose electrophoretic lanes. Our method has been shown to yield comparable accuracy with manual lane tracking results, and to successfully process 98% of DNA fingerprinting gels with no human intervention. Note to Practitioners-The work described here increases the efficiency of a high-throughput laboratory environment where many electrophoretic gels are run. Results of these runs are captured by a digital imager, and constitute the primary data for the downstream analysis. Ideally, unsupervised bioinformatic approaches can be depended upon for accurate and precise data extraction from the image, which can then be used for research such as physical genome mapping, and genomic rearrangement discovery. The first task in the image analysis is to identify the regions on the gel that the sample has traveled through, referred to here as lane tracking. Previously, manual inspection of the lane tracking results for each gel was necessary to ensure accurate data extraction. The time dedicated to this task in the laboratory could have been better spent completing less automatable tasks. The completion of this work frees the technician from checking on the software results. Finally, the software is fitted with quality control heuristics which can select any particularly nonconforming gels for manual inspection. This project is designed to be applicable to similar laboratory experiments. However, to achieve the highly dependable results we describe here, some setup time might be required for data prepared outside our institution.