wizaRdry (version 0.2.6)

sift: Filter data frame by superkey parameters, rows, and columns

Description

Filter data frame by superkey parameters, rows, and columns

Usage

sift(
  df,
  rows = NULL,
  cols = NULL,
  record_id = NULL,
  src_subject_id = NULL,
  subjectkey = NULL,
  site = NULL,
  subsiteid = NULL,
  sex = NULL,
  race = NULL,
  ethnic_group = NULL,
  phenotype = NULL,
  phenotype_description = NULL,
  status = NULL,
  lost_to_followup = NULL,
  twins_study = NULL,
  sibling_study = NULL,
  family_study = NULL,
  sample_taken = NULL,
  visit = NULL,
  week = NULL,
  arm = NULL,
  interview_date = NULL
)

Value

A filtered dataframe based on the provided parameters, and containing only the columns specified in 'cols'. If no columns are specified, returns the entire dataframe with applied row filters.

Arguments

df

Dataframe to be filtered and trimmed based on the provided parameters.

rows

Optional; either a single row name or a vector of row names to be retained in the final output. If NULL or empty, all rows in the dataframe are retained.

cols

Optional; either a single column name or a vector of column names to be retained in the final output. If NULL or empty, all columns in the dataframe are retained.#' Data Filter

record_id

Optional; either a single record_id or a vector of record_ids to filter the dataframe by

src_subject_id

Optional; either a single subject ID or a vector of subject IDs to filter the dataframe by

subjectkey

Optional; either a single subjectkey or a vector of subjectkeys to filter the dataframe by

site

Optional; either a single site value or a vector of site values to filter the dataframe by (e.g., Yale, NU)

subsiteid

Optional; either a single subsiteid or a vector of subsiteids to filter the dataframe by

sex

Optional; either a single sex value or a vector of sex values at birth to filter the dataframe by (e.g., 'M', 'F')

race

Optional; either a single race value or a vector of race values to filter the dataframe by

ethnic_group

Optional; either a single ethnic_group value or a vector of ethnic_group values to filter the dataframe by

phenotype

Optional; either a single phenotype value or a vector of phenotype values to filter the dataframe by

phenotype_description

Optional; either a single phenotype_description or a vector of phenotype_descriptions to filter the dataframe by

status

Optional; either a single status string or a vector of status conditions to filter the dataframe by. Used if either 'state' or 'status' column exists in the dataframe. Can include values like 'complete', 'completed baseline', 'completed 12m', 'completed 24m', etc.

lost_to_followup

Optional; either a single value or a vector of values to filter the dataframe by (checks both 'lost_to_followup' and 'lost_to_follow-up' columns)

twins_study

Optional; either a single twins_study value or a vector of twins_study values to filter the dataframe by

sibling_study

Optional; either a single sibling_study value or a vector of sibling_study values to filter the dataframe by

family_study

Optional; either a single family_study value or a vector of family_study values to filter the dataframe by

sample_taken

Optional; either a single sample_taken value or a vector of sample_taken values to filter the dataframe by

visit

Optional; either a single visit value or a vector of visit values to filter the dataframe by. Only used if 'visit' column exists in the dataframe.

week

Optional; either a single week value or a vector of week values to filter the dataframe by. Only used if 'week' column exists in the dataframe.

arm

Optional; either a single arm value or a vector of arm values to filter the dataframe by (e.g., drug, placebo)

interview_date

Optional; can be either: - A date string in various formats (ISO, US, etc.) to filter data up to that date - A boolean TRUE to return only rows with non-NA interview_date values

Examples

# Create a sample dataframe
sample_df <- data.frame(
  record_id = c("REC001", "REC002", "REC003", "REC004"),
  src_subject_id = c("SUB001", "SUB002", "SUB003", "SUB004"),
  subjectkey = c("KEY001", "KEY002", "KEY003", "KEY004"),
  site = c("Yale", "NU", "Yale", "NU"),
  phenotype = c("A", "B", "A", "C"),
  visit = c(1, 2, 2, 1),
  state = c("complete", "completed baseline", "in progress", NA),
  status = c(NA, NA, NA, "complete"),
  lost_to_followup = c(FALSE, FALSE, TRUE, NA),
  interview_date = c("2023-01-15", "2023/02/20", NA, "2023-03-10")
)

# Set row names for demonstration
rownames(sample_df) <- c("foo", "bar", "baz", "qux")

# Filter by specific date
filtered1 <- sift(sample_df, 
                 cols = c("src_subject_id", "phenotype"), 
                 visit = 2, 
                 interview_date = "01/31/2023")
                 
# Filter to include only rows with non-NA interview dates
filtered2 <- sift(sample_df, 
                 interview_date = TRUE)
                 
# Filter by status (works with either state or status column)
filtered3 <- sift(sample_df,
                 status = c("complete", "completed baseline"))
                 
# Filter with specific row names
filtered4 <- sift(sample_df,
                 rows = c("foo", "qux"))
                 
# Filter with vector of visit values
filtered6 <- sift(sample_df,
                 visit = c(1, 2))
                 
# Filter by lost_to_followup
filtered10 <- sift(sample_df,
                 lost_to_followup = FALSE)
                 
# Filter by src_subject_id
filtered11 <- sift(sample_df,
                 src_subject_id = c("SUB001", "SUB004"))
                 
# Multiple filters combined
filtered12 <- sift(sample_df,
                 site = "Yale",
                 visit = 1,
                 cols = c("record_id", "src_subject_id", "site"))