This document shows I have processed all the refDB data into a data.table or data.frame. The reason to do so is to reduce the computational time for read-in.

Statistics

I plan first to format the data as data.table, which uses two keys.

## [1] "AA" "SS"
## [1] "stats.CA"
##    AA SS Stat Value
## 1:  A  A   mu 53.44
## 2:  A  A   sd  1.91
## 3:  A  B   mu 51.53
## 4:  A  B   sd  1.48
## 5:  A  C   mu 52.84
## 6:  A  C   sd  1.64
## [1] "stats.CB"
##    AA SS Stat Value
## 1:  A  A   mu 53.44
## 2:  A  A   sd  1.91
## 3:  A  B   mu 51.53
## 4:  A  B   sd  1.48
## 5:  A  C   mu 52.84
## 6:  A  C   sd  1.64
## [1] "stats.cov"
##    AA SS      Value
## 1:  A  A -2.1356194
## 2:  A  B -0.9933715
## 3:  A  C -0.5770540
## 4:  A  H -0.3102361
## 5:  B  A -0.6925940
## 6:  B  B  1.0453800

RefDB Data

## ALA 
## "A"
##          Obs Label AA Atom Type Value       ID
##       1:   1     2  S   HA    H  4.49 bmr10002
##       2:   2     2  S  HB2    H  3.81 bmr10002
##       3:   3     2  S  HB3    H  3.81 bmr10002
##       4:   5     2  S   CA    C 58.35 bmr10002
##       5:   6     2  S   CB    C 63.79 bmr10002
##      ---                                      
## 1968051: 478    81  N   HA    H  4.48   bmr979
## 1968052: 479    81  N  HB2    H  2.71   bmr979
## 1968053: 480    81  N  HB3    H  2.68   bmr979
## 1968054: 481    81  N HD21    H  7.56   bmr979
## 1968055: 482    81  N HD22    H  6.79   bmr979