Research scientists, medical professionals, and the academic community publish their findings every year culminating in time series publication numbers of data that form nonlinear trends over time. Understanding these trends would allow researchers to predict future levels of need and interest in specific research areas within their discipline. The problem with studying these trends is defining exactly what their quantitative behavior will be in the future. Trends in publication frequency can be described by plotting sub-discipline publication numbers over time. In this study, we assign specific sigmoidal equations to each sub-discipline studied by doing a Boolean search of PubMed for publication numbers on research topics related to six molecules, ten cell types, and four organ types all related to immunology. Our approach was to transform the original data by reduction of the x-axis and then curve fit the original data set to the best fitting curve which could be analyzed by non-linear regression. This technique was essential to arriving at an accurate prediction of the expected number of publications. Our findings are immunological publication numbers of cells, molecules, and organ types in immunology have exhibited significant trends that give R2 values higher than 0.95 and that in our areas of study only sigmoidal trend behaviors are observed. We propose that demonstrated trends in publications counts will be informative to researchers allowing prediction of growth of interest in their respective fields of study in immunology. Also, we affirm that any predictions made from our research can be verified by chi-square analysis. Keywords: time series, publication numbers, publication frequency, sigmoidal, immunology, prediction