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$index)
##      [,1] [,2] [,3] [,4] [,5] [,6] [,7] [,8] [,9] [,10]
## [1,] 6048 9885  353 2971 8945 6637 4376 4840 7113  4829
## [2,] 4566 5533 4546 6554 3587 6773  781 6464 1773  7421
## [3,] 9301 4308 5739  571 5126 4764 9939 1072 8372  5958
## [4,] 9181 5582 2810 7584 8324 7026 6776 7403  489  2008
## [5,] 5176 4487 5412 9882 9251 3958 1164 8756 8020  7579
## [6,] 5306  271 9964 8753 7782 9840 8334 8626  638  6371
head(fout$distance)
##           [,1]      [,2]      [,3]      [,4]      [,5]      [,6]     [,7]
## [1,] 0.8753423 1.0324337 1.0401479 1.0433857 1.0487950 1.0555128 1.072550
## [2,] 0.7456681 0.9882122 1.0138134 1.0305669 1.0346680 1.0355890 1.043177
## [3,] 1.0490947 1.0645473 1.0732491 1.0743933 1.0768465 1.1141785 1.150830
## [4,] 0.8340625 0.8681348 0.8935820 0.8972260 0.9046458 0.9814676 1.020612
## [5,] 0.8079657 0.8583915 0.8963977 0.9582708 0.9645010 0.9685468 1.029258
## [6,] 1.0281247 1.0578119 1.0763279 1.0804365 1.0964895 1.0981416 1.099325
##          [,8]     [,9]    [,10]
## [1,] 1.077113 1.088926 1.090039
## [2,] 1.065462 1.070953 1.089345
## [3,] 1.155441 1.171077 1.176814
## [4,] 1.025402 1.026696 1.027654
## [5,] 1.030429 1.030792 1.042128
## [6,] 1.105819 1.107315 1.128199

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$index)
##      [,1] [,2] [,3] [,4] [,5]
## [1,] 2588 9668 3292  428 2745
## [2,] 9731 8783 9113  217 5817
## [3,] 4088 2983 1205 9533 9397
## [4,] 2249 4930 4691 3497  789
## [5,] 6888 4084 7679 5406 6175
## [6,] 4421 8209 9718 5691 9212
head(qout$distance)
##           [,1]      [,2]      [,3]      [,4]      [,5]
## [1,] 0.8495727 0.8568265 0.9252666 0.9386955 0.9692541
## [2,] 0.9696178 1.0059217 1.0248015 1.0293496 1.0342718
## [3,] 0.7496009 0.8444536 0.9318359 0.9610884 0.9643283
## [4,] 0.8002510 0.9452562 0.9696701 1.0036349 1.0458778
## [5,] 0.9468699 0.9619154 0.9694815 0.9710934 0.9857361
## [6,] 0.9074824 0.9646798 1.0710313 1.0750747 1.1089975

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] "/tmp/Rtmpk2FJHs/filee6283700ca92.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-apple-darwin15.6.0 (64-bit)
## Running under: OS X El Capitan 10.11.6
## 
## Matrix products: default
## BLAS:   /Library/Frameworks/R.framework/Versions/3.6/Resources/lib/libRblas.0.dylib
## LAPACK: /Library/Frameworks/R.framework/Versions/3.6/Resources/lib/libRlapack.dylib
## 
## locale:
## [1] C/en_US.UTF-8/en_US.UTF-8/C/en_US.UTF-8/en_US.UTF-8
## 
## 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