Drinking water companies are responsible for producing and supplying safe, high quality water for costumers. The drinking water quality should meet the standard at the customer tap, therefore the supplier is forced to analyze the water quality not only at the outlet from the treatment plant but in the distribution network as well. The aim of this study was identification and discrimination of microorganisms in water samples from different points of distribution network in Riga, Latvia. The classification of bacteria is generally based on the morphology and biochemical reactions of the bacteria. These measurements are time consuming, requiring training and expertise. Infrared (IR) spectroscopic measurements of bacteria followed by a formal chemometrics analysis could offer advantages of speed and consistency [1]. Since the amount of pathogenic micro-organisms in water is low, after one or more purification processes, concentration is necessary to detect and quantify these organisms. Simmons [2] describes a cross flow ultra filter (Hemoflow-filter) that is used for the concentration of microorganisms in water. An important advantage of this filter is that it concentrates parasitic protozoa, bacteria, spores and viruses/phages. This study was performed on drinking water from drinking water plants from Riga, used as water sources surface water and groundwater. Here we present results of concentrated drinking water samples which were investigated by Fourier Transform Infrared (FT-IR) spectroscopy. Presence of all typical peaks of main biochemical cell components: lipids – 2928 – 2856 cm-1 (CH2, CH3 simetric/asimetric stretching vibrations); proteins - Amide I and II – 1549 and 1655 cm-1 ; nucleic acids – 1242 cm-1 (P=O), and carbohydrates - 1080 cm-1 (C-O-C, C-O, C-O-H, P=O of PO2-, valent stretching vibrations of COC groups and ring vibration modes in the composition of cyclic structures) verified biomass while qualitative and quantitative differences of components indicated various strains or their mix/consortium. Results of both analytical methods showed that water samples contain significant amount of biomass. We found distinct difference in the FT-IR spectra of biomass isolated from treated groundwater or surface water. In some Hemoflow samples Fluorescence in Situ Hybridization (FISH) method indicated E. coli while Pseudomonas fluorescens was identified in all samples. This study showed that combination of biomass separation and FT-IR spectral analysis is a rapid and simple approach for characterization of microorganisms in drinking water.