Position Weight Matrix or Acyclic Probabilistic Finite Automaton: Which model to use? A decision rule inferred for the prediction of transcription factor binding sites.

Prediction of transcription factor binding sites (TFBS) is an example of application of Bioinformatics where DNA molecules are represented as sequences of A, C, G and T symbols. The most used model in this problem is Position Weight Matrix (PWM). Notwithstanding the advantage of being simple, PWMs cannot capture dependency between nucleotide positions, which may affect prediction performance. Acyclic Probabilistic Finite Automata (APFA) is an alternative model able to accommodate position dependencies. However, APFA is a more complex model, which means more parameters have to be learned. In this paper, we propose an innovative method to identify when position dependencies influence preference for PWMs or APFAs. This implied using position dependency features extracted from 1106 sets of TFBS to infer a decision tree able to predict which is the best model - PWM or APFA - for a given set of TFBSs. According to our results, as few as three pinpointed features are able to choose the best model, providing a balance of performance (average precision) and model simplicity.

Assessing the infection risk of Giardia and Cryptosporidium in public drinking water delivered by surface water systems in Sao Paulo State, Brazil.

A survey of Giardia and Cryptosporidium was conducted in surface water used as drinking water sources by public water systems in four densely urbanized regions of Sao Paulo State, Brazil. A Quantitative Microbial Risk Assessment, based on protozoa concentrations, was performed to estimate the probability of protozoa infection associated with drinking water ingestion. A total of 206 source water samples were analyzed over a 24 month period using the USEPA Method 1623. The risk of infection was estimated using an exponential dose response model, children and adults exposure and a gamma distribution for (oo)cyst concentrations with three scenarios for treating censored data. Giardia was detected in 102 of the samples, and 19 of them were also positive for Cryptosporidium, with maximum concentrations of 97.0 cysts/L and 6.0 oocysts/L, respectively. Risk distributions were similar for the three scenarios. In the four regions, the estimated risk of Giardia infection per year, for adults and children, ranged from 0.29% to 2.47% and from 0.08% to 0.70%, respectively. Cryptosporidium risk infection varied from 0.15% to 0.29% for adults and from 0.04% to 0.08% for children. In both cases, the calculated risk surpassed the risk of infection of 10(-4) (1:10,000) defined as tolerable by USEPA for a yearly exposure. The probability of Giardia infection was very close to the rates of acute diarrheic disease for adults (1% to 3%) but lower for children (2% to 7%). The daily consumption of drinking water was an important contributing factor for these differences. The Microbiological Risk Assessment carried out in this study provides an indication of infection risks by Giardia and Cryptosporidium in the population served by these source waters. Strategies for source water protection and performance targets for the water treatment should be established to achieve the required level of public health risk.