1 Introduction

The BiocNeighbors package provides several algorithms for approximate neighbor searches:

  • The Annoy (Approximate Nearest Neighbors Oh Yeah) method uses C++ code from the RcppAnnoy package. It works by building a tree where a random hyperplane partitions a group of points into two child groups at each internal node. This is repeated to construct a forest of trees where the number of trees determines the accuracy of the search. Given a query data point, we identify all points in the same leaf node for each tree. We then take the union of leaf node sets across trees and search them exactly for the nearest neighbors.
  • The HNSW (Hierarchical Navigable Small Worlds) method uses C++ code from the RcppHNSW package. It works by building a series of nagivable small world graphs containing links between points across the entire data set. The algorithm walks through the graphs where each step is chosen to move closer to a given query point. Different graphs contain links of different lengths, yielding a hierarchy where earlier steps are large and later steps are small. The accuracy of the search is determined by the connectivity of the graphs and the size of the intermediate list of potential neighbors.

These methods complement the exact algorithms described previously. Again, it is straightforward to switch from one algorithm to another by simply changing the BNPARAM argument in findKNN and queryKNN.

2 Identifying nearest neighbors

We perform the k-nearest neighbors search with the Annoy algorithm by specifying BNPARAM=AnnoyParam().

nobs <- 10000
ndim <- 20
data <- matrix(runif(nobs*ndim), ncol=ndim)

fout <- findKNN(data, k=10, BNPARAM=AnnoyParam())
head(fout$index)
##      [,1] [,2] [,3] [,4] [,5] [,6] [,7] [,8] [,9] [,10]
## [1,] 8771 5059   56 4070 7404 2231 8376 8101 3327  5661
## [2,] 9923 2309 9300 6519 4514 2897 1081 5905 1752  6343
## [3,]  780 6729 5792 8257 2741  367 6741  767  788   963
## [4,] 4786 5471 1459 5984  187 7896 9861 9790 3155  5987
## [5,] 6601 1948 3652 1731 1278 7916 2014 2056 6725   142
## [6,] 1586 2850 6887  149 7567 7155 3651  964 5058  8658
head(fout$distance)
##           [,1]      [,2]      [,3]      [,4]      [,5]      [,6]      [,7]
## [1,] 0.8715133 0.9678817 0.9833923 0.9893116 1.0315957 1.0330536 1.0331842
## [2,] 0.9367853 1.0039316 1.0237066 1.0342082 1.0561885 1.0673904 1.0785345
## [3,] 0.9123796 0.9180802 0.9481109 0.9552479 0.9676833 0.9927770 0.9976734
## [4,] 0.7858602 0.8895646 0.8904961 0.9430057 0.9458280 0.9496274 0.9524800
## [5,] 0.9369186 0.9492112 0.9567255 0.9793873 0.9822933 0.9842297 0.9861553
## [6,] 0.8238962 0.9035017 0.9434627 1.0584267 1.1050088 1.1274583 1.1323702
##           [,8]      [,9]    [,10]
## [1,] 1.0371625 1.0666491 1.082250
## [2,] 1.0863938 1.0982471 1.099542
## [3,] 1.0026765 1.0033996 1.019348
## [4,] 0.9617489 0.9933506 1.000173
## [5,] 0.9936376 1.0054382 1.021920
## [6,] 1.1386094 1.1392430 1.141825

We can also identify the k-nearest neighbors in one dataset based on query points in another dataset.

nquery <- 1000
ndim <- 20
query <- matrix(runif(nquery*ndim), ncol=ndim)

qout <- queryKNN(data, query, k=5, BNPARAM=AnnoyParam())
head(qout$index)
##      [,1] [,2] [,3] [,4] [,5]
## [1,] 2016  973 4386 8287 1124
## [2,]  499 9522 9925   73 5016
## [3,] 5965 6174 4464  260 6317
## [4,] 5192 3421 9886 3100 9923
## [5,]   31 2993 3642 3055  346
## [6,] 3702 7485 5121 8620 9426
head(qout$distance)
##           [,1]      [,2]      [,3]      [,4]      [,5]
## [1,] 0.8418320 0.9973949 1.0008547 1.0193620 1.0248270
## [2,] 0.8564282 0.9597957 1.0455211 1.0805972 1.0828418
## [3,] 0.7865698 0.8234340 0.9240150 0.9528975 0.9762501
## [4,] 0.9860422 1.0171522 1.0203296 1.0233701 1.0313755
## [5,] 0.8818297 0.9705501 0.9707427 1.0054229 1.0084831
## [6,] 0.8956466 0.8967537 0.9626351 0.9702065 0.9805890

It is similarly easy to use the HNSW algorithm by setting BNPARAM=HnswParam().

3 Further options

Most of the options described for the exact methods are also applicable here. For example:

  • subset to identify neighbors for a subset of points.
  • get.distance to avoid retrieving distances when unnecessary.
  • BPPARAM to parallelize the calculations across multiple workers.
  • BNINDEX to build the forest once for a given data set and re-use it across calls.

The use of a pre-built BNINDEX is illustrated below:

pre <- buildIndex(data, BNPARAM=AnnoyParam())
out1 <- findKNN(BNINDEX=pre, k=5)
out2 <- queryKNN(BNINDEX=pre, query=query, k=2)

Both Annoy and HNSW perform searches based on the Euclidean distance by default. Searching by Manhattan distance is done by simply setting distance="Manhattan" in AnnoyParam() or HnswParam().

Users are referred to the documentation of each function for specific details on the available arguments.

4 Saving the index files

Both Annoy and HNSW generate indexing structures - a forest of trees and series of graphs, respectively - that are saved to file when calling buildIndex(). By default, this file is located in tempdir()1 On HPC file systems, you can change TEMPDIR to a location that is more amenable to concurrent access. and will be removed when the session finishes.

AnnoyIndex_path(pre)
## [1] "C:\\Users\\biocbuild\\bbs-3.10-bioc\\tmpdir\\Rtmp4UPzyy\\file12b859ed4484.idx"

If the index is to persist across sessions, the path of the index file can be directly specified in buildIndex. This can be used to construct an index object directly using the relevant constructors, e.g., AnnoyIndex(), HnswIndex(). However, it becomes the responsibility of the user to clean up any temporary indexing files after calculations are complete.

5 Session information

sessionInfo()
## R version 3.6.2 (2019-12-12)
## Platform: x86_64-w64-mingw32/x64 (64-bit)
## Running under: Windows Server 2012 R2 x64 (build 9600)
## 
## Matrix products: default
## 
## locale:
## [1] LC_COLLATE=C                          
## [2] LC_CTYPE=English_United States.1252   
## [3] LC_MONETARY=English_United States.1252
## [4] LC_NUMERIC=C                          
## [5] LC_TIME=English_United States.1252    
## 
## attached base packages:
## [1] stats     graphics  grDevices utils     datasets  methods   base     
## 
## other attached packages:
## [1] BiocNeighbors_1.4.2 knitr_1.28          BiocStyle_2.14.4   
## 
## loaded via a namespace (and not attached):
##  [1] Rcpp_1.0.3          bookdown_0.17       lattice_0.20-40    
##  [4] digest_0.6.25       grid_3.6.2          stats4_3.6.2       
##  [7] magrittr_1.5        evaluate_0.14       rlang_0.4.4        
## [10] stringi_1.4.6       S4Vectors_0.24.3    Matrix_1.2-18      
## [13] rmarkdown_2.1       BiocParallel_1.20.1 tools_3.6.2        
## [16] stringr_1.4.0       parallel_3.6.2      xfun_0.12          
## [19] yaml_2.2.1          compiler_3.6.2      BiocGenerics_0.32.0
## [22] BiocManager_1.30.10 htmltools_0.4.0