Skip to main content

Table 3 Prediction on archived GOA files using TO (see Eq. (5))

From: Interspecies gene function prediction using semantic similarity

   MicroAvgF1 MacroAvgF1 1-RankLoss Fmax RAccuracy
CC H →H 0.8374 0.7212 0.8968 0.8729 0.1773
  M →H 0.8351 0.7241 0.8969 0.8743 0.1762
  M+H →H 0.8586 0.7641 0.9771 0.8751 0.2845
  D →H 0.8351 0.7322 0.8941 0.8693 0.1662
  D+H →H 0.8512 0.7476 0.9422 0.8654 0.2469
  A →H 0.8317 0.6982 0.8832 0.8860 0.1488
  A+H →H 0.8366 0.7223 0.8962 0.8726 0.1732
  M →M 0.7765 0.6075 0.8826 0.8526 0.2122
  H →M 0.7805 0.6130 0.8836 0.8295 0.2166
  H+M →M 0.8132 0.6547 0.9597 0.8665 0.3418
  D →M 0.7726 0.6142 0.8659 0.8320 0.2092
  D+M →M 0.7993 0.6357 0.9252 0.8384 0.2928
  A →M 0.7758 0.6278 0.8700 0.8324 0.2105
  A+M →M 0.7770 0.6088 0.8807 0.8447 0.2142
MF H →H 0.8569 0.8228 0.9293 0.8952 0.1687
  M →H 0.8542 0.8213 0.9262 0.8941 0.1527
  M+H →H 0.8711 0.8382 0.9763 0.9064 0.2510
  D →H 0.8524 0.8348 0.9413 0.8717 0.1426
  D+H →H 0.8606 0.8349 0.9588 0.8979 0.1901
  A →H 0.8456 0.8225 0.9124 0.8941 0.1026
  A+H →H 0.8535 0.8181 0.9260 0.8933 0.1489
  M →M 0.7756 0.6946 0.8985 0.8692 0.1697
  H →M 0.7804 0.6957 0.9096 0.8569 0.1677
  H+M →M 0.7851 0.7104 0.9374 0.8806 0.2051
  D →M 0.7695 0.6811 0.8963 0.8602 0.1538
  D+M →M 0.7851 0.7082 0.9356 0.8731 0.2051
  A →M 0.7635 0.6941 0.8816 0.8588 0.1249
  A+M →M 0.7752 0.6840 0.8993 0.8616 0.1683
BP H →H 0.8460 0.8019 0.9605 0.8729 0.2472
  M →H 0.8428 0.7998 0.9586 0.7818 0.2316
  M+H →H 0.8500 0.8071 0.9745 0.8751 0.2664
  D →H 0.8385 0.8036 0.9605 0.7901 0.2101
  D+H →H 0.8443 0.8016 0.9613 0.7877 0.2387
  A →H 0.8314 0.7943 0.9333 0.7591 0.1755
  A+H →H 0.8389 0.7933 0.9520 0.7813 0.2120
  M →M 0.7960 0.7101 0.9519 0.7813 0.2405
  H →M 0.7980 0.7073 0.9532 0.7767 0.2481
  H+M →M 0.8024 0.7163 0.9677 0.8244 0.2643
  D →M 0.7886 0.7137 0.9508 0.7756 0.2129
  D+M →M 0.7832 0.7059 0.9410 0.7765 0.2318
  A →M 0.7795 0.7020 0.9158 0.7373 0.1791
  A+M →M 0.7723 0.6923 0.9326 0.7716 0.2106
  1. H →H directly uses GO annotations of Human to predict annotations of Human genes. M →H only employs annotations of genes from Mouse to predict annotations of Human genes. M+H →H uses GO annotations of genes from Mouse and Human to predict annotations of Human genes. D+H →H uses annotations of genes from Danio rerio and Human to predict annotations of Human genes. A+H →H uses annotations of genes from Arabidopsis thaliana and Human to predict annotations of Human genes. M →M, H+M →M, D+M →M and A+M →M follow the similar protocol, but predict annotations of Mouse genes. The data in boldface is the statistically significant best among these comparing methods for a particular target species, and the significance is checked by paired t-test at 95% confidence intervals