Fluctuation Programs

Below are the m-files and a description of the scripts from the Lang and Murray 2008 Genetics paper. Greg Lang

findMLm

Input: data from fluctuation assay.

Output: most likely value of m given the data. Command: findMLm(data)

This program requires the following programs to run: scoreData, generateLD.

findMLmTwoParam

Input: data from fluctuation assay.

Output: most likely values of m and d given the data. Command: findMLmTwoParameter(data)

This program requires the following programs to run: scoreDataTwoParam, generateLD, generatePO, generateTwoParam.

scoreData

Input: data from fluctuation assay and m.

Output: -log probability of observing the data given m. Command: scoreData(data, m)

This program requires the following program to run: generateLD.

scoreDataTwoParam

Input: data from fluctuation assay, m, and d.

Output: -log probability of observing data given m and d. Command: scoreDataTwoParam(data, m, d)

This program requires the following programs to run: generateLD, generatePO, generateTwoParam.

generateLD

Input: m, and max.

Output: The Luria-Delbrück distribution from 0 to max with parameter m. Command: generateLD(m, max)

generatePO

Input: lambda, and max.

Output: The Poisson distribution from 0 to max with parameter lambda. Command: generatePO(lambda, max)

generateTwoParam

Input: m, d, and max.

Output: The combined Luria-Delbrück and Poisson distribution from 0 to

max with parameters m and lambda = m x d. Command: generateTwoParam(m, d, max)

This program requires the following programs to run: generateLD, generatePO.

sampleLD

Input: m, number of samples.

Output: Random variables from the Luria-Delbrück distribution with

parameter m. Command: sampleLD(m, samples)

This program requires the following program to run: generateLD.

sampleTwoParam

Input: m, number of samples.

Output: Random variables from a combined Luria-Delbrück and Poisson

distribution with parameters m and lambda = m x d. Command: sampleTwoParam(m, d, samples)

This program requires the following programs to run: generateLD, generatePO, generateTwoParam.

SSDScoreLD

Input: data from fluctuation assay.

Output: sum of the squared differences between the cumulative distribution

of the data and the cumulative Luria-Delbrück distribution with

parameter m determined from the data. Command: SSDScore(data)

This program requires the following programs to run: findMLm, scoreData, generateLD.

SSDScoreTwoParam

Input: data from fluctuation assay.

Output: sum of the squared differences between the cumulative distribution

of the data and the cumulative combined Luria-Delbrück and Poisson distribution with parameters m and lambda = m x d determined from the data.

Command: SSDScoreTwoParam(data)