trueFalseSum: summary of true positive and false negative assignments for...

In WMBEdmands/compMS2Miner: an automatable metabolite identification, visualization and data-sharing R package for high-resolution LC-MS datasets

Description

summary of true positive and false negative assignments for each score metric

Usage

trueFalseSum(object, ...)

Arguments

object	a "compMS2" class object
n	integer the number of top annotations to consider (default = 5). i.e. if the correct structure is found in the top 5 structures then it is considered a true positive.
minSimScore	numeric chemical similarity score (values between 0 and 1). This score is used to identify the compound contained in the metID comments table (default = 1). The is performed rather than by name as there can be multiple hits of the same structure in the top annotations. To check if the metric is in roughly the right ballpark highly similar structures to the correct annotations can be considered by setting the argument to a lower Tanimoto chemical similarity score (e.g. 0.9).
specDbOnly	logical if TRUE (default) then only the spectral database annotations are used to calculate the true positive and false negative rates. If FALSE then all possible_identities in the metID comments table are used.
verbose	logical if TRUE display progress bars.

Details

The weighted average takes in to account the number of possible candidates by weighting the mean correct rank by the number of possible candidates. e.g. In this way a rank of 1 out of 10 possible structures will be ranked lower than a rank of 3 out of 1,000 possible structures. The latter is a more important/impressive reflection of ranking ability of than the former. If a more extensive chemical database is utilized such as PubMed compound for example then a far larger number of candidates may be considered than a much smaller curated database such as HMDB.

Value

a list summarizing the true positive and false negative outputs, a summary plot of the results and also weighted average rank for each score metric.

WMBEdmands/compMS2Miner documentation built on May 9, 2019, 10:04 p.m.