BIOSTAT 214: Data Management and Advanced R Programming (Winter 2020))
Section outline
-
-
This online book covers the material discussed in Biostat214 and will be updated weekly
-
Lecture: Introduction to Data Science & the R language
- Data Science in biomedicine: goals and challenges
- The R language and why it’s great for biomedical data
- The R package ecosystem: CRAN, Bioconductor, GitHub
- The RStudio Integrated Development Environment (IDE)
- Rmarkdown, Jupyter notebooks, RStudio notebooks
- Base R vs. “tidyverse” vs. data.table
Location: Mission Hall 1407 -
Lecture: Data input/output, data types & structures, indexing, flow control
- Reading data from CSV files, Excel spreadsheets
- R data structures and data types; type conversions
- Types: logical, integer, double, complex, character
- Structures: vector, matrix, array, list, data.frame
- Factors
- class(), typeof(), str()
- Indexing: filter cases, select variables
- bracket notation
- subset(), split()
- Logical vs integer indexing; which
- Long <-> wide data conversions
- Saving data to .csv, .rds, .RData, .xlsx
- Logical operations
Faculty: Stathis Gennatas
Location: Mission Hall 1407 -
Lecture: Flow control; Data summarization, aggregation; vectorization; functional programming
- Flow control: for loops, if-then-else, while
- Summarize statistics
- Aggregate data
- Vectorized operations
- apply(), lapply(), sapply(), vapply(), tapply(), mapply()
- by(), subset(), aggregate()
- Reduce, Filter, Find, Position, Map, Negate
- The pipe operator %>%
- Functions in R
- Functions as first class objects
- do.call()
Faculty: Stathis Gennatas
Location: Mission Hall 1407 -
-
-
Lecture: Writing functions; Environments & scoping; Writing documentation; Performance profiling
- Writing your own functions to streamline your data processing pipelines.
- R environments and scoping.
- Reading (others’) R code.
- Writing documentation using roxygen2.
- Profiling code
Faculty: Stathis Gennatas
Location: Mission Hall 1407 -
Lecture: Data preprocessing, imputation, dimensionality reduction; table joins
- Handling missing values: imputation, last observation carried forward.
- Categorical data: nominal vs ordinal
- Standardization / feature scaling
- Dimensionality reduction for high dimensional inputs.
- Merging data sources: (SQL-type) join operations between tables: inner, full/left/right outer joins.
Faculty: Stathis Gennatas
Location: Mission Hall 1406 (Please note room changed) -
-
-
Lecture: Efficient data analysis with data.table
Faculty: Stathis Gennatas
Location: Zoom onlyZoom Link: https://stanford.zoom.us/j/875634241 or +1 650 724 9799 (Meeting ID: 875 634 241)
-
Watch Media Resource
-
Faculty: Stathis Gennatas
Lecture: Course Review, Q&A, project presentations
Presentations & handing in of completed projects
Location: Zoom onlyZoom Link: https://stanford.zoom.us/j/875634241 or +1 650 724 9799 (Meeting ID: 875 634 241)
-
Watch Media Resource
