Data.frame two list, each list contain different length of elements

Question

I have a data frame with two lists of variables. Each observation in the list contains different length of elements. For example the 4th of the variable “accession” contains one element but 7th contains two elements. current dataframe

I want to make a new data frame combine two lists together which looks like: final dataframe I want

Thanks for helping me!

This is data frame I am currently having.

library(rentrez)


search <- entrez_search(db="gds", term=paste0("disease", " AND gse[ETYP]") , retMax = 15) 
id <- unlist(search$ids)
UID <- c(sapply(id, paste0, collapse=""))
pub.summary <- entrez_summary(db = "gds", id = UID ,  
                              always_return_list = TRUE)
summary <- extract_from_esummary(esummaries = pub.summary , 
                                           elements = c("samples"),
                                           simplify = T)
df <- data.frame(summary)
df <-data.frame(t(df))
df <- df %>% mutate()
df

This is the data frame result I wish to have

#  accession                                  title
#1 GSM3955152                                Cancer3
   GSM3955155                              Adjacent3
   GSM3955757 SW480 cells, HES1-binding RNAs/LncRNAs
   GSM3955153                              Adjacent1
   GSM3955150                                Cancer1
   GSM3955151                                Cancer2
#2 GSM33026213                      his4wk_sensitized_uti_1
   GSM3302681                         3his4wk_resolved_pbs_2
   GSM3302624                           c57bl6j_pbs_9
.
.
.
.
#4 GSM3955757                      SW480 cells, HES1-binding RNAs/LncRNAs
.
.
.
.
#15 GSM3934992                    control rep4 [N_0039]
    GSM3935006                    control rep15 [W_010]
    GSM3935012                    control rep17 [W_023]
    GSM3934989                    control rep1 [N_0026]
END
 
    

Answer 1

Update

Based on the OP's updates, an option is to specify simplify = FALSE in the extract_from_esummary to return as list, then extract the first list element fom each list and rbind to create a single dataframe

summary <- extract_from_esummary(esummaries = pub.summary , 
                                           elements = "samples",
                                           simplify = FALSE)


out <- do.call(rbind, lapply(summary, `[[`, 1))
row.names(out) <- NULL
head(out)
#  accession                                  title
#1 GSM3955152                                Cancer3
#2 GSM3955155                              Adjacent3
#3 GSM3955757 SW480 cells, HES1-binding RNAs/LncRNAs
#4 GSM3955153                              Adjacent1
#5 GSM3955150                                Cancer1
#6 GSM3955151                                Cancer2

---

An option would be pad the list elements with NA to keep the length same in both columns (if one is of different length) and then unnest

library(dplyr)
library(purrr)
df1 %>%
   mutate(n = pmax(lengths(accession), lengths(title))) %>% 
   mutate_at(vars(accession, title), ~ 
         map2(., n, ~ `length<-`(.x, .y))) %>% 
   select(-n) %>%
   unnest(cols = c(accession, title))
# A tibble: 12 x 2
#   accession title
#   <chr>     <chr>
# 1 A         a    
# 2 B         b    
# 3 C         c    
# 4 <NA>      d    
# 5 <NA>      e    
# 6 A         a    
# 7 B         b    
# 8 C         c    
# 9 D         <NA> 
#10 E         <NA> 
#11 A         d    
#12 B         <NA> 

---

Or an option is to gather into 'long' format, then unnest the 'val' column and spread it back to 'wide' format

library(tidyr)
df1 %>%
    mutate(rn = row_number()) %>% 
    gather(key, val, -rn) %>%
    unnest(val) %>%
    group_by(rn, key) %>% 
    mutate(i1 = row_number()) %>% 
    spread(key, val) %>% 
    ungroup %>% 
    select(-rn, -i1)

data

df1 <- tibble(accession = list(LETTERS[1:3], LETTERS[1:5], LETTERS[1:2]), 
       title = list(letters[1:5], letters[1:3], letters[4]))