The sum of a number of measurements divided by the total number of measurements.
What is the Mean?
This type of distribution is for counted data. The data is counted, not measured.
What is a discrete probability distribution?
Used in Bayesian statistical inference, this distribution expresses one's beliefs about this quantity before some evidence is taken into account.
What is a Prior Distribution?
This mass parameter value is expressed in RFU.
A mathematical process that uses random number generation and acceptance-rejection sampling to obtain a sampling from the posterior distribution of a set of variables (mass parameters for STRmix).
What is Markov Chain Monte Carlo?
The number which appears most often in a group of numbers/measurements.
What is the Mode?
This type of data is measured and can take any value within a given range.
What is continuous data?
This type of distribution is generated after the MCMC trials numerous combinations of parameter values to describe the observed data. Each free parameter (mass parameters) will have this type of distribution.
What is the Posterior distribution?
This mass parameter is expressed as the change in peak height over the base pair size of the allele =
ln(rfu)/bp
What is the Degradation rate?
This algorithm is an MCMC method for obtaining a sequence of random samples from a probability distribution that may otherwise be difficult to sample from.
What is the Metropolis Hastings (accept/reject) algorithm?
The number which falls in the middle of a group of sorted numbers/measurements.
What is the median?
This type of probability distribution is for continuous variables/data.
What is a Continuous distribution?
This type of probability is the probability that an event will happen after all evidence or background information has been taken into account.
What is the Posterior probability?
Posterior probability = prior probability + new evidence (called likelihood).
This parameter is dimensionless,. It is a measure of the efficiency of the amplification at each locus.
What is LSAE (Locus Specific Amplification Efficiency)?
The % of time the “worse” model is accepted is determined by the probability of the proposed divided by the probability of the current guess. It is a comparison of where the model has been compared to the new model.
What are the Accept-Reject criteria (Metropolis Hastings)?
A measure of how spread out the numbers are from the mean. This value is represented by sigma σ.
What is a standard deviation?
This distribution is bell-shaped where the data tends to be around a central value (mean) with no bias left or right.
What is a Normal (Gaussian) Distribution?
This type of probability is the probability an event will happen before the evidence has been taken into account.
What is the Prior probability?
Posterior probability = prior probability + new evidence (called likelihood).
This mass parameter is dimensionless. It describes the variation between replicate amplifications of the sample. Scales peaks up or down between replicates. It is represented by Ry.
What is the Replicate Multiplier?
At each iteration the MCMC will be sitting on a particular set of values for the mass parameters and peak height variance constants (c2 and K2). This term describes the next move it makes.
What is stepping?
A measure of how spread out the numbers are. This value is represented by sigma squared:
σ 2
What is the Variance?
This continuous distribution is used by STRmix when sampling allele and stutter variance values. This distribution is determined by alpha and beta which were determined by Model Maker during validation.
What is a Gamma Distribution?
STRmix uses this function when comparing observed and expected peak heights (rather than probabilities). It can do this because it is not the absolute value for weights that are important but rather their size relative to each other.
What is a Probability Density?
The area under the curve gives the probability. The height of the curve at any point is the probability density.
This mass parameter is dimensionless. It describes the variation between replicate amplifications of the sample with different PCR kits. It is represented by Bk.
What is the Kit Multiplier?
When proposing new values for the next MCMC iteration, the values will be chosen close to the current set of values. The step distance is determined by this.
What is the Gaussian Random Walk?
The distance that the step-size can be is based on a normal distribution with a mean equal to the current parameter value and a variance that dictates step-size.