Detecting spatial patterns in biological array experiments

J Biomol Screen. 2003 Aug;8(4):393-8. doi: 10.1177/1087057103254282.


Chemical genetic screening and DNA and protein microarrays are among a number of increasingly important and widely used biological research tools that involve large numbers of parallel experiments arranged in a spatial array. It is often difficult to ensure that uniform experimental conditions are present throughout the entire array, and as a result, one often observes systematic spatially correlated errors, especially when array experiments are performed using robots. Here, the authors apply techniques based on the discrete Fourier transform to identify and quantify spatially correlated errors superimposed on a spatially random background. They demonstrate that these techniques are effective in identifying common spatially systematic errors in high-throughput 384-well microplate assay data. In addition, the authors employ a statistical test to allow for automatic detection of such errors. Software tools for using this approach are provided.

Publication types

  • Research Support, Non-U.S. Gov't
  • Research Support, U.S. Gov't, P.H.S.

MeSH terms

  • Analysis of Variance
  • Data Interpretation, Statistical
  • Fourier Analysis
  • Gene Expression Profiling / methods*
  • Genetic Techniques*
  • Molecular Probe Techniques*
  • Oligonucleotide Array Sequence Analysis*
  • Protein Array Analysis
  • Quality Control
  • Reproducibility of Results
  • Sensitivity and Specificity