Filter a comparison to include only the best hits for each query
best_hit.RdTakes a comparison object and filters it to include only the best hit
for each query. The query names are provided by group_by, and the best hit
is determined by sorting each set of queries based on a set of columns
provided by sort_order.
Usage
best_hit(
comparison,
group_by = "name1",
sort_order = c("bit_score", "aln_len", "per_id", "direction")
)Arguments
- comparison
A
comparisonobject to filter.- group_by
A character string referring to a column in
comparisonthat holds the query names.- sort_order
A character vector of column names that determine what the best hit is, in order of importance. The first column is used to determine the best hits per query, and each subsequent column is only used in case of ties.
Details
Designed to find the best hits per query from comparison objects
generated from tabular BLAST or DIAMOND results. For each column from
sort_order, the maximum (or minimum) value is taken for each unique query
name from group_by, in the order provided by sort_order, until each query
only 1 hit left. The maximum values are taken for each sort_order column,
unless the column provided is one of: "mism", "gaps", "e_value",
"name1", "name2", "start1", "start2", "end1", or "end2".
Examples
## Read example blastp results
infile <- system.file('extdata/blastp_example1.tab', package = 'genoPlotR')
## comparison before filtering for best hits
blast_comparison <- read_comparison_from_blast(infile)
print_comparison(blast_comparison)
#> start1 end1 start2 end2 name1 name2 per_id aln_len
#> <int> <int> <int> <int> <char> <char> <num> <int>
#> 1: 117 502 1 385 HKIDPCCJ_00001 LABPAKCO_00001 58.7 387
#> 2: 1 376 1 379 HKIDPCCJ_00002 LABPAKCO_00002 56.5 379
#> 3: 19 651 12 645 HKIDPCCJ_00004 LABPAKCO_00005 74.6 634
#> 4: 23 655 8 644 HKIDPCCJ_00004 LABPAKCO_01957 51.4 642
#> 5: 11 811 10 812 HKIDPCCJ_00005 LABPAKCO_00006 62.8 803
#> 6: 11 810 6 809 HKIDPCCJ_00005 LABPAKCO_01958 40.0 817
#> 7: 1 172 1 190 HKIDPCCJ_00007 LABPAKCO_00009 55.8 197
#> 8: 1 172 1 164 HKIDPCCJ_00007 LABPAKCO_02342 50.0 174
#> 9: 93 535 67 509 HKIDPCCJ_00029 LABPAKCO_00168 28.7 453
#> 10: 87 521 107 546 HKIDPCCJ_00029 LABPAKCO_00977 26.3 452
#> 11: 328 529 335 539 HKIDPCCJ_00029 LABPAKCO_01459 33.2 205
#> 12: 88 534 100 548 HKIDPCCJ_00029 LABPAKCO_01460 23.7 451
#> 13: 304 498 312 502 HKIDPCCJ_00029 LABPAKCO_02399 31.3 201
#> 14: 328 521 28 227 HKIDPCCJ_00029 LABPAKCO_02526 29.8 208
#> 15: 328 519 4 214 HKIDPCCJ_00029 LABPAKCO_01272 29.1 213
#> 16: 1 637 1 644 HKIDPCCJ_00034 LABPAKCO_00103 56.1 645
#> 17: 2 210 11 270 HKIDPCCJ_00034 LABPAKCO_01390 31.5 260
#> 18: 8 303 12 290 HKIDPCCJ_00034 LABPAKCO_01699 30.1 306
#> 19: 100 531 78 511 HKIDPCCJ_00038 LABPAKCO_00168 24.1 456
#> 20: 250 513 267 544 HKIDPCCJ_00038 LABPAKCO_00977 26.1 284
#> 21: 267 513 290 533 HKIDPCCJ_00038 LABPAKCO_01460 27.6 254
#> 22: 325 503 331 513 HKIDPCCJ_00038 LABPAKCO_02399 28.1 185
#> 23: 109 473 105 460 HKIDPCCJ_00038 LABPAKCO_01612 22.5 374
#> 24: 359 501 368 516 HKIDPCCJ_00038 LABPAKCO_01459 29.6 152
#> 25: 320 514 22 226 HKIDPCCJ_00038 LABPAKCO_02526 25.8 209
#> 26: 323 485 2 187 HKIDPCCJ_00038 LABPAKCO_01272 22.1 204
#> 27: 451 514 331 398 HKIDPCCJ_00038 LABPAKCO_00394 32.4 68
#> 28: 8 338 4 334 HKIDPCCJ_00047 LABPAKCO_00854 54.1 333
#> 29: 4 336 1 333 HKIDPCCJ_00047 LABPAKCO_02361 27.1 336
#> 30: 1 442 1 443 HKIDPCCJ_00048 LABPAKCO_01328 33.8 444
#> 31: 1 438 1 440 HKIDPCCJ_00048 LABPAKCO_02752 32.4 442
#> 32: 1 379 1 378 HKIDPCCJ_00049 LABPAKCO_00839 72.5 382
#> 33: 15 348 9 310 HKIDPCCJ_00049 LABPAKCO_00753 32.2 342
#> start1 end1 start2 end2 name1 name2 per_id aln_len
#> mism gaps e_value bit_score direction
#> <int> <int> <num> <num> <num>
#> 1: 157 2 1.32e-153 440.0 1
#> 2: 162 1 1.18e-140 401.0 1
#> 3: 160 1 0.00e+00 946.0 1
#> 4: 298 11 9.19e-211 601.0 1
#> 5: 297 1 0.00e+00 989.0 1
#> 6: 460 7 4.55e-185 547.0 1
#> 7: 55 2 2.90e-56 172.0 1
#> 8: 75 2 8.61e-50 154.0 1
#> 9: 303 10 3.22e-49 174.0 1
#> 10: 304 13 2.30e-31 124.0 1
#> 11: 134 3 3.10e-30 120.0 1
#> 12: 338 6 4.45e-29 117.0 1
#> 13: 122 4 2.08e-22 97.1 1
#> 14: 124 6 2.37e-14 69.7 1
#> 15: 128 7 3.15e-14 71.2 1
#> 16: 274 5 7.15e-256 715.0 1
#> 17: 127 4 8.69e-31 124.0 1
#> 18: 177 10 5.54e-24 102.0 1
#> 19: 300 9 3.02e-28 114.0 1
#> 20: 184 8 3.67e-21 93.2 1
#> 21: 167 6 5.92e-19 86.3 1
#> 22: 125 2 9.99e-18 82.4 1
#> 23: 263 7 2.73e-17 80.9 1
#> 24: 95 3 1.32e-12 66.2 1
#> 25: 137 6 9.21e-11 58.9 1
#> 26: 100 7 6.56e-05 41.6 1
#> 27: 42 2 2.91e-04 39.7 1
#> 28: 149 3 6.03e-110 322.0 1
#> 29: 239 4 5.43e-29 111.0 1
#> 30: 291 3 2.02e-80 252.0 1
#> 31: 293 4 7.28e-74 235.0 1
#> 32: 98 3 5.42e-194 537.0 1
#> 33: 184 11 4.93e-43 149.0 1
#> mism gaps e_value bit_score direction
## Filter for best hits and print results
bh_comparison <- best_hit(blast_comparison)
print_comparison(bh_comparison)
#> start1 end1 start2 end2 name1 name2 per_id aln_len
#> <int> <int> <int> <int> <char> <char> <num> <int>
#> 1: 117 502 1 385 HKIDPCCJ_00001 LABPAKCO_00001 58.7 387
#> 2: 1 376 1 379 HKIDPCCJ_00002 LABPAKCO_00002 56.5 379
#> 3: 19 651 12 645 HKIDPCCJ_00004 LABPAKCO_00005 74.6 634
#> 4: 11 811 10 812 HKIDPCCJ_00005 LABPAKCO_00006 62.8 803
#> 5: 1 172 1 190 HKIDPCCJ_00007 LABPAKCO_00009 55.8 197
#> 6: 93 535 67 509 HKIDPCCJ_00029 LABPAKCO_00168 28.7 453
#> 7: 1 637 1 644 HKIDPCCJ_00034 LABPAKCO_00103 56.1 645
#> 8: 100 531 78 511 HKIDPCCJ_00038 LABPAKCO_00168 24.1 456
#> 9: 8 338 4 334 HKIDPCCJ_00047 LABPAKCO_00854 54.1 333
#> 10: 1 442 1 443 HKIDPCCJ_00048 LABPAKCO_01328 33.8 444
#> 11: 1 379 1 378 HKIDPCCJ_00049 LABPAKCO_00839 72.5 382
#> mism gaps e_value bit_score direction
#> <int> <int> <num> <num> <num>
#> 1: 157 2 1.32e-153 440 1
#> 2: 162 1 1.18e-140 401 1
#> 3: 160 1 0.00e+00 946 1
#> 4: 297 1 0.00e+00 989 1
#> 5: 55 2 2.90e-56 172 1
#> 6: 303 10 3.22e-49 174 1
#> 7: 274 5 7.15e-256 715 1
#> 8: 300 9 3.02e-28 114 1
#> 9: 149 3 6.03e-110 322 1
#> 10: 291 3 2.02e-80 252 1
#> 11: 98 3 5.42e-194 537 1