Selecting rows with values within /-1 of each other in R-CodePudding

I am trying to do some spectral analysis and am aiming to isolate spectral peaks that are within /-1 MZ values of each other. I have made an R script to combine all of my spectral information into one data frame with all possible peptides, but it is a very large data frame and I am trying to find a way to streamline finding tandem peaks. I have no experience with R or any programming language so any help would be appreciated.

I have tried by making a column that finds the difference in MZ values between adjacent rows and then filtered by looking for those with only a value of 1 but this causes me to miss the first/last peak in the tandem spectra.

Here is an example of what a portion of my data frame looks like where the diff column is

mutate(diff = Mz_Round - lag(Mz_Round))

And then filtered to only include where diff == 1.

 precursorMz Mz_Round HW Intensity Reg Intensity diff
136    256.6814   251.15         2108          2305   NA
137    256.6814   255.18         6491          3910   NA
138    256.6814   255.68         2292          1114   NA
139    256.6814   260.20        43010         23230   NA
140    256.6814   261.20         9452          6388    1
141    256.6814   262.19         6440          3487   NA

For this specific example, I want to extract rows 139 and 140 because they are both within 1 mz unit of each other but if I were to filter solely by which rows have a diff value of 1 then row 139 would be missing.

CodePudding user response：

data <-
  data.frame(
    Mz_Round = c(
      251.15,
      255.18,
      255.68,
      260.20,
      261.20,
      262.19
    )
  )

data |>
  dplyr::mutate(
    diff_1 = Mz_Round - dplyr::lag(Mz_Round),
    diff_2 = dplyr::lead(Mz_Round) - Mz_Round 
  ) |>
  dplyr::filter(
    diff_1 == 1 | diff_2 == 1
  )
#>   Mz_Round diff_1 diff_2
#> 1    260.2   4.52   1.00
#> 2    261.2   1.00   0.99

^{Created on 2022-11-08 with reprex v2.0.2}