A comprehensive understanding of factors that influence microbial competition and cooperation, their diversity and processes will be greatly beneficial in many research areas. Current tools for microflora determinations are far from suitable for high-throughput monitoring of development in complex microbial communities. Here, we describe the application of a calibration free method, multivariate curve resolution with alternating least squares (MCR-ALS), for identification and quantification of different microbes in mixture samples. The idea is to utilize MCR-ALS to enable close monitoring of ecology in a variety of microbial communities. The data from two designed experiments consisting of DNA sequence spectra measured on mixtures were analysed with MCR-ALS using no prior information on the data except for appropriate constraints, such as non-negativity and closure. The results were compared both to the known true concentrations as well as to the results obtained from the well-established multivariate calibration method partial least squares (PLS) regression. MCR-ALS performed as well as PLS regression, successfully extracting all pure bacterial spectra and quantitative information on these, with 97.81% and 97.91% explained variance for the first and the second data set, respectively. Copyright © 2008 John Wiley & Sons, Ltd.