1 Introduction

The SMAD (Statistical Modelling of AP-MS Data) package is designed to process Affinity Purification-Mass Spectrometry (AP-MS) data. Its primary goal is to compute confidence scores that help researchers distinguish true protein-protein interactions (PPI) from non-specific background contaminants.

In a typical AP-MS experiment, many proteins might be identified, but only a fraction are bona fide interactors. SMAD implements several validated statistical models to assign probability scores to these interactions.

2 Installation

You can install the SMAD package from Bioconductor using the following command:

if (!require("BiocManager", quietly = TRUE))
    install.packages("BiocManager")

BiocManager::install("SMAD")

3 Preparing Input Data

SMAD requires input data as a dataframe. The standard format includes identifiers for the experiment run, the bait protein, the prey protein, and quantitative measurements (like spectral counts and protein length).

3.1 Example Dataset

We provide a sample dataset TestDatInput, which is a subset of the BioPlex 2.0 data focusing on apoptosis-related proteins.

library(SMAD)
#> Loading required package: RcppAlgos
data("TestDatInput")
head(TestDatInput)
#>       idRun idBait idPrey countPrey lenPrey
#> 7452  68982  TIMP2  ACTC1        15     377
#> 8016  66491  CASP1   CDK4         9     303
#> 7162  68486   BTG3  RPL24         3     157
#> 8086  66491  CASP1 IMPDH2         9     514
#> 23653 72934    LUM  THOP1         7     689
#> 9196  67747    FAS   RFC5         9     340

The columns required for most scoring functions are:

Column Name	Description
`idRun`	Unique ID for the AP-MS experiment run
`idBait`	Identifier for the bait protein used in the pull-down
`idPrey`	Identifier for the identified prey protein
`countPrey`	Quantitative measure (e.g., peptide or spectral counts)
`lenPrey`	Length of the prey protein (used for normalization)

4 Scoring Protein Interactions

SMAD offers multiple scoring algorithms. This guide focuses on two popular methods: CompPASS and HGScore.

4.1 CompPASS (Comparative Proteomic Analysis Software Suite)

CompPASS is a “spoke model” algorithm that identifies high-confidence interactors by comparing occurrences across a large number of experiments. It was originally developed by Sowa et al. (2009) and widely used in the BioPlex projects.

The output includes several metrics, with the WD-score (Weighted D-score) being the most commonly used for ranking.

# Run CompPASS scoring
scoreCompPASS <- CompPASS(TestDatInput)

# View the top results
head(scoreCompPASS[order(scoreCompPASS$scoreWD, decreasing = TRUE), ])
#>      idBait idPrey AvePSM     scoreZ   scoreS   scoreD Entropy  scoreWD
#> 1082   CD69 SPTAN1    101   7.938223 81.02469 81.02469       0 3.798496
#> 4618    TNF ATP2A2     37   7.938223 49.04080 49.04080       0 2.299068
#> 1169   CDK2 CREBBP     30   7.938223 44.15880 44.15880       0 2.070197
#> 3536   MMP2   MMP2     58 233.709532 43.41659 43.41659       0 2.035279
#> 1161   CDK2  CCNB1     27   7.938223 41.89272 41.89272       0 1.963961
#> 1519   DAP3 MRPS35     26   7.938223 41.10961 41.10961       0 1.927248

We can visualize the distribution of scores to see the separation of high-confidence interactors:

par(mfrow = c(1, 1))
plot(sort(scoreCompPASS$scoreWD, decreasing = TRUE), 
     pch = 20, col = "royalblue",
     xlab = "Ranked Interactions", 
     ylab = "WD-score",
     main = "CompPASS WD-score Distribution")
abline(h = mean(scoreCompPASS$scoreWD) + 2 * sd(scoreCompPASS$scoreWD), 
       col = "red", lty = 2)
legend("topright", legend = "Mean + 2SD", col = "red", lty = 2)

4.2 HGScore (Hypergeometric Score)

HGScore is based on a hypergeometric distribution error model (Hart et al., 2007), incorporating Normalized Spectral Abundance Factor (NSAF) to account for protein length. Unlike CompPASS, HGScore can incorporate a “matrix model” perspective, often leading to a larger number of inferred interactions.

# Run HG scoring
scoreHG <- HG(TestDatInput)

# View the top results
head(scoreHG[order(scoreHG$HG, decreasing = TRUE), ])
#>       InteractorA InteractorB ppiTN tnA tnB           PPI NMinTn       HG
#> 44384      CDKN1A      CDKN1B    13 403 396 CDKN1A~CDKN1B 477317 41.28085
#> 39530       CCNB1       CKS1B     9 212 212   CCNB1~CKS1B 477317 39.24069
#> 39557       CCNB1        SKP2     7 212 208    CCNB1~SKP2 477317 30.00500
#> 49442       CKS1B        SKP2     7 212 208    CKS1B~SKP2 477317 30.00500
#> 39568       CCND2      CDKN1A     7 195 403  CCND2~CDKN1A 477317 25.38601
#> 44500      CDKN1B        SKP2     7 396 208   CDKN1B~SKP2 477317 25.01539

Visualizing the HGScore distribution:

plot(sort(scoreHG$HG, decreasing = TRUE), 
     pch = 20, col = "darkorange",
     xlab = "Ranked Interactions", 
     ylab = "HGscore",
     main = "HGScore Distribution")

5 Advanced Scoring and Further Reading

While CompPASS and HGScore are excellent starting points, SMAD includes several other advanced scoring methods such as:

SAINTexpress: A widely adopted Bayesian framework for AP-MS.
PE (Purification Enrichment): A Bayesian classifier combining spoke and matrix models.
DICE and Hart: Specialized scores for prey-prey interaction affinity.

For a detailed showcase of all these functions, please refer to the Scoring Functions in SMAD vignette:

vignette("scoring_functions", package = "SMAD")

6 References

Appendix

A Session Information

sessionInfo()
#> R version 4.6.0 alpha (2026-04-05 r89794)
#> Platform: x86_64-pc-linux-gnu
#> Running under: Ubuntu 24.04.4 LTS
#> 
#> Matrix products: default
#> BLAS:   /home/biocbuild/bbs-3.23-bioc/R/lib/libRblas.so 
#> LAPACK: /usr/lib/x86_64-linux-gnu/lapack/liblapack.so.3.12.0  LAPACK version 3.12.0
#> 
#> locale:
#>  [1] LC_CTYPE=en_US.UTF-8       LC_NUMERIC=C              
#>  [3] LC_TIME=en_GB              LC_COLLATE=C              
#>  [5] LC_MONETARY=en_US.UTF-8    LC_MESSAGES=en_US.UTF-8   
#>  [7] LC_PAPER=en_US.UTF-8       LC_NAME=C                 
#>  [9] LC_ADDRESS=C               LC_TELEPHONE=C            
#> [11] LC_MEASUREMENT=en_US.UTF-8 LC_IDENTIFICATION=C       
#> 
#> time zone: America/New_York
#> tzcode source: system (glibc)
#> 
#> attached base packages:
#> [1] stats     graphics  grDevices utils     datasets  methods   base     
#> 
#> other attached packages:
#> [1] SMAD_1.27.5      RcppAlgos_2.10.0 BiocStyle_2.39.0
#> 
#> loaded via a namespace (and not attached):
#>  [1] vctrs_0.7.3         cli_3.6.6           knitr_1.51         
#>  [4] magick_2.9.1        rlang_1.2.0         xfun_0.57          
#>  [7] otel_0.2.0          purrr_1.2.2         generics_0.1.4     
#> [10] jsonlite_2.0.0      data.table_1.18.2.1 glue_1.8.1         
#> [13] htmltools_0.5.9     tinytex_0.59        sass_0.4.10        
#> [16] gmp_0.7-5.1         rmarkdown_2.31      tibble_3.3.1       
#> [19] evaluate_1.0.5      jquerylib_0.1.4     fastmap_1.2.0      
#> [22] yaml_2.3.12         lifecycle_1.0.5     bookdown_0.46      
#> [25] BiocManager_1.30.27 compiler_4.6.0      dplyr_1.2.1        
#> [28] pkgconfig_2.0.3     Rcpp_1.1.1-1        tidyr_1.3.2        
#> [31] digest_0.6.39       R6_2.6.1            tidyselect_1.2.1   
#> [34] pillar_1.11.1       magrittr_2.0.5      bslib_0.10.0       
#> [37] withr_3.0.2         tools_4.6.0         cachem_1.1.0

1. Introduction to SMAD

2026-04-17

Package