Classifying classic books using the nltk (natural language toolkit) package in Python. Tag text with parts-of-speech (POS), chunk by POS patterns (e.g., adjective followed by noun), create subtrees based on chunking patterns. Count tables by POS (numpy package), plotting (matplotlib).
ANCOVA mixed models in SAS Procedure Mixed, which uses REML estimation (restricted maximum likelihood). Type III tests of fixed effects, allowing multiple-degree interactions. Nested random effects. Null hypothesis testing. Standardize covariates to z-score (i.e., mean 0, standard deviation 1). Repeated measures. A priori contrasts.
Cadney_et_al_2021_Effects_of_early-life_Western_diet_and_voluntary_exercise_on_adult_mice.pdf
Schmill_et_al_2020_conditioned_place_preference_for_cocaine_and_methylphenidate_in_HR_mice.pdf
Hiramatsu_and_Garland_2018_high-runner_mice_conserve_more_fat_despite_increased_exercise.pdf
Hiramatsu_et_al_2017_maternal_exposure_to_Western_diet_affects_offspring.pdf
Heritability estimates by mixed "animal" models in ASReml-R and MCMCglmm in R: A type of mixed model that partitions phenotypic variation and estimates direct additive genetic variance, indirect additive genetic variance, and their covariance. 95% Confidence Intervals based on fixed effects +/- standard error (proLik from nadiv package in R).
Dewan_et_al_2019_Indirect_Genetic_Effects_on_Voluntary_Wheel-Running_Behavior.pdf
Hiramatsu_and_Garland_2016_Nature_or_Nurture-Heritability_in_the_Classroom.pdf
Complex tree modeling in Onyx.
Hiramatsu_and_Garland_2018_high-runner_mice_conserve_more_fat_despite_increased_exercise.pdf
Java-based GATK3 and GATK4 (genome analysis toolkits) best practices pipelines on the command line. Imputation of missing data, cleaning of noisy data.