Yates algorithm

The factorial approach to the design of experiments allows all the tests involving several factors to be combined in the calculation of the main effects and their interactions. For a 23 design, there are 3 main effects, 3 two-factor interactions, and 1 three-factor interaction. Yates algorithm can be used to determine the main effects and their interactions (17). The data can also be represented as a multiple linear regression model... 

Historically, factorial designs were introduced by Sir R. A. Fisher to counter the then prevalent idea that if one were to discover the effect of a factor, all other factors must be held constant and only the factor of interest could be varied. Fisher showed that all factors of interest could be varied simultaneously, and the individual factor effects and their interactions could be estimated by proper mathematical treatment. The Yates algorithm and its variations are often used to obtain these estimates, but the use of least squares fitting of linear models gives essentially identical results. 

The Yates algorithm is a formal procedure for estimating the P s for full two-level factorial designs [Yates (1936)]. The Yates algorithm is related to the fast Fourier transform. We describe the Yates algorithm here, and illustrate it s use for the 1 full factorial design discussed in Section 14.2. 

In the early days of experimental design the Yates algorithm made hand calculations easier and minimized calculational errors. However, similar results can be obtained with modem regression analysis packages. 

Frakes, W. B. and Baeza-Yates, R. (eds.) (1992) Information retrieval data structures and algorithms. Prentice Hall, Englewood Cliffs, NJ. 

Two algorithms are available to perform all the calculations in a very simple way, namely the Box, Hunter and Hunter (BH ) algorithm and Yates s algorithm. Both are considered below for a typical and simple example of a 2 factorial design. Assume we are studying the influence of pH (A), temperature (B) and time (C) over the yield (response in %) of the extraction of a metal from a complex analytical matrix, just before conducting the extracts to an ICP device. The levels of each factor, fixed by the analyst, are pH (A), 3 (—), 5(4-) temperature (B), 40 (-), 60 C ( + ) and time (C), 1 (-), 2h. (- -). The matrix design and the experimental data are as follows ... 

As can be seen, all the elements of the extra column I are + , so that the respective effect that they evaluate is just the arithmetic mean of the responses, named in experimental design Mean effect. Although the BI P algorithm has many uses, the most rapid way to calculate effects (particularly when we consider many factors) is by means of an algorithm developed by Yates. 

Yates s algorithm (named after Frank Yates, a co-worker of Ronald Fisher, 1902-94) is applied to the observations after they have been arranged in the standard order. As shown in Table 2.2, the Yates calculations start by evaluating as many auxiliary columns as factors are considered, in our example three columns El, E2 and E3 for a 2 design. 

The Yates algorithm is easily generalized for any 2k FUFE. For k factor A, B, C,... 

MacCoss MJ, Wu CC, Liu H, Sadygov R, Yates JR 3rd. A correlation algorithm for the automated quantitative analysis of shotgun proteomics data. Anal Chem 2003 75 6912-6921. 

The effects were calculated using Yates algorithm. Table 2 summarizes the effect analysis for density of dried aerogel. The level of experimental design at low temperature, pressure and autoclave volume was denoted while more severe conditions were + . 

A method often used for hand-calculation of effects from factorial designs is the Yates algorithm.[2] As this algorithm also assumes that the levels of the variables are exactly as specified by the design. It is therefore suggested that it should not be used to the evaluation of synthesis experiments. 

Where Xe is the value to be codified, X is the average value of the defined limits for each variable and Xmax the maximum limit of the variable. From the results of these evaluations by use of the algorithm developed by Yates [6] employing the State-ease "Design Expert (5.0.9)" programme, the effects of each variable were calculated and plotted against a half normal percent probability, shown in Fig. 1 where those variables with little or no effect on the NO conversion value fall on the line. Variables that cause the greatest effect over the NO conversion are those that are located away from the line. 

A method developed by Eng and Yates uses a MS/MS cross-correlation algorithm called SEQUEST . - In this method, an unknown protein is digested using trypsin and the resulting peptides are used to collect product MS-MS scans. These MS-MS scans are used to search a protein database for a match within the expected MS-MS data from the known tryptic peptides. 

MacCoss, M.J. Wu, C.C. Yates, J.R. Probability-based validation of protein identifications using a modified SEQUEST algorithm. Anal. Chem. 2002, 74, 5593-5599. 

These calculations have been presented in some detail in order to make the principles clear. An algorithm due to Yates (see Bibliography) simplifies the calculation. 

---