added supplementary info

supplementary_info/DatasetAdditionalInfo.MD

@@ -0,0 +1,22 @@
+# Additional Information on BDB and KIBA
+
+The following figure illustrates the binding score distributions and quartiles in the BDB and KIBA datasets. A strong peak at $pK_d = 5$ is observed for BDB, because $K_d$ of weak interactions are often recorded as $K_d \geq 10000$ $(pK_d \leq 5)$.
+
+![](../img/binding_score_distributions.png)
+
+We create train/validation/test splits out of BDB and KIBA with warm and cold biomolecules (see the manuscript for more details). We report the average number of proteins, ligands, and interactions in the training and test sets in the following table, alongside standard deviations in the parentheses.
+
+| Dataset | Fold         | #Proteins            | #Ligands               | #Interactions              |
+| ------- | ------------ | -------------------- | ---------------------- | -------------------------- |
+| BDB     | Train        | 403.4    $\pm$  2.8  | 740.8    $\pm$  19.46  | 17988.2     $\pm$  646.45  |
+| BDB     | Validation   | 355.0    $\pm$  5.62 | 170.0    $\pm$  11.05  | 1494.2      $\pm$  56.17   |
+| BDB     | Warm         | 354.4    $\pm$  3.44 | 179.6    $\pm$  5.28   | 1494.4      $\pm$  56.32   |
+| BDB     | Cold Ligand  | 376.0    $\pm$  4.38 | 84.8    $\pm$  5.53    | 2448.8      $\pm$  373.48  |
+| BDB     | Cold Protein | 43.6     $\pm$  2.15 | 264.8    $\pm$  90.17  | 2360.0      $\pm$  216.02  |
+| BDB     | Cold Both    | 41.4     $\pm$  3.07 | 30.8    $\pm$  11.92   | 274.6      $\pm$  36.19    |
+| KIBA    | Train        | 200.6    $\pm$  1.36 | 1834.6   $\pm$  6.41   | 77264.4     $\pm$  814.94  |
+| KIBA    | Validation   | 193.0    $\pm$  1.67 | 1467.2   $\pm$  23.75  | 6650.2      $\pm$  69.53   |
+| KIBA    | Warm         | 192.0    $\pm$  3.16 | 1476.2   $\pm$  17.7   | 6650.6      $\pm$  69.1    |
+| KIBA    | Cold Ligand  | 193.0    $\pm$  2.45 | 140.0    $\pm$  5.59   | 6810.0      $\pm$  570.52  |
+| KIBA    | Cold Protein | 14.6     $\pm$  0.8  | 1296.0   $\pm$  179.09 | 6259.6      $\pm$  1024.25 |
+| KIBA    | Cold Both    | 14.0     $\pm$  1.1  | 100.2    $\pm$  14.55  | 468.6      $\pm$  37.89    |

supplementary_info/OutOfDatasetAllModels.csv

@@ -0,0 +1,36 @@
+,,Accuracy,,Precision,,Recall,,F1,,AUC
+Train Dataset,Model,Mean,Std,Mean,Std,Mean,Std,Mean,Std,Mean
+KIBA,DeepDTA,0.829022021,0.010749762,0.31040456,0.020405,0.207502987,0.03479197,0.24583721,0.021436183,0.566980563
+KIBA,DeepDTA - IDDTA (BD),0.829209845,0.006798421,0.293646927,0.022751665,0.185663082,0.033530546,0.226221088,0.029410768,0.557881176
+KIBA,DeepDTA- IDDTA (BG),0.819203368,0.004696735,0.28080571,0.019095087,0.214384707,0.02384578,0.242792593,0.020989417,0.56420303
+KIBA,DeepDTA- BOW (BD),0.827467617,0.010356973,0.28936218,0.032332599,0.187574671,0.044727455,0.225127479,0.037074155,0.55767945
+KIBA,DeepDTA- BOW (BG),0.810207254,0.007810392,0.257553875,0.020803436,0.213811231,0.035102976,0.232655882,0.027643565,0.558758022
+KIBA,BPE-DTA,0.820906736,0.01226448,0.246859466,0.044139528,0.156224612,0.038263846,0.189819025,0.039769338,0.540667091
+KIBA,BPEDTA - BOW (BG),0.832130829,0.009500644,0.249915869,0.042490612,0.117228196,0.028154841,0.158205887,0.031830479,0.530717488
+KIBA,BPEDTA - BOW (BD),0.818212435,0.015839599,0.246086189,0.037590463,0.155316607,0.016840031,0.187883322,0.007881915,0.538725919
+KIBA,BPEDTA - ID (BD),0.816431347,0.032021931,0.276476198,0.04315266,0.190824373,0.051231744,0.217334936,0.018326673,0.552666354
+KIBA,BPEDTA - ID (BG),0.826612694,0.016590142,0.271283117,0.074091025,0.156224612,0.032526657,0.197424105,0.045041716,0.543967335
+KIBA,LMDTA,0.81634715,0.009150259,0.303630638,0.015367183,0.272640382,0.034951903,0.285978618,0.019046148,0.587112474
+KIBA,LMDTA - IDDTA (BG),0.81382772,0.007708506,0.29278691,0.021013592,0.261792115,0.009454156,0.276196339,0.013206293,0.581081485
+KIBA,LMDTA - IDDTA (BD),0.82117228,0.008211465,0.309126278,0.015802444,0.256630824,0.033439151,0.279056895,0.019134225,0.583153397
+KIBA,LMDTA - BOW' (BG),0.822279793,0.008080088,0.311151727,0.026659405,0.254719235,0.02648991,0.279486627,0.023109181,0.582988013
+KIBA,LMDTA - BOW' (BD),0.818335492,0.007590701,0.308900209,0.014085704,0.273548387,0.025252427,0.289368158,0.01581499,0.58864533
+KIBA,LMDTA - BOW (BG),0.818490933,0.006508503,0.305905171,0.028832375,0.268243728,0.03146096,0.285659899,0.029642493,0.586498713
+KIBA,LMDTA - BOW (BD),0.814540155,0.003815256,0.290545894,0.006510072,0.256917563,0.032546312,0.271955048,0.020943972,0.579438366
+BDB,DeepDTA,0.787630344,0.005322511,0.324392076,0.015503491,0.095480376,0.020490006,0.146098922,0.025017741,0.523897415
+BDB,DeepDTA - BOW (BD),0.788132093,0.003046999,0.316115467,0.032871381,0.090525662,0.022331726,0.139984596,0.029676764,0.522320064
+BDB,DeepDTA - BOW (BG),0.776805188,0.00546601,0.300606632,0.017693696,0.121266205,0.018750223,0.171997934,0.019856819,0.527022302
+BDB,DeepDTA - IDDTA (BG),0.779684273,0.010054804,0.311045065,0.035959851,0.11433138,0.019286912,0.165401376,0.019655804,0.52616195
+BDB,DeepDTA - IDDTA (BD),0.78951105,0.003299523,0.33762716,0.014986693,0.098117563,0.009174978,0.15172671,0.010967684,0.526066366
+BDB,BPE-DTA,0.782595785,0.008711753,0.321500027,0.005088482,0.117190552,0.03826057,0.168094919,0.039912033,0.529053519
+BDB,BPEDTA - BOW (BG),0.729294308,0.011021099,0.270343267,0.010754048,0.23977979,0.029651623,0.253067034,0.018887832,0.542772451
+BDB,BPEDTA - BOW (BD),0.771176722,0.003764361,0.292450595,0.006949342,0.134532055,0.016049787,0.183817636,0.01588394,0.52859322
+BDB,BPEDTA - IDDTA (BG),0.741467702,0.031915291,0.276033598,0.02498593,0.198907832,0.062318656,0.222589872,0.03780359,0.534733742
+BDB,BPEDTA - IDDTA (BD),0.76672583,0.018065987,0.290455981,0.012984762,0.14359794,0.04846431,0.186226212,0.042462659,0.529292685
+BDB,LMDTA ,0.753366328,0.012544957,0.247825388,0.012854873,0.139904102,0.036970096,0.176400275,0.0311559,0.519616133
+BDB,LMDTA - BOW' (BG),0.753929516,0.012580932,0.249518539,0.026278389,0.139877464,0.031137645,0.17774806,0.029058738,0.519954577
+BDB,LMDTA - BOW' (BD),0.760559775,0.005629733,0.257506138,0.022613454,0.132525306,0.027657369,0.174199562,0.029060809,0.521257055
+BDB,LMDTA - BOW (BG),0.766973291,0.005335952,0.259211652,0.02450048,0.114056118,0.011179712,0.158293749,0.014887451,0.518189405
+BDB,LMDTA - BOW (BD),0.765848622,0.00512969,0.265809323,0.01238758,0.124791334,0.020280877,0.169142773,0.021052123,0.52158376
+BDB,LMDTA - IDDTA (BG),0.75421111,0.012421243,0.25086088,0.021671619,0.143127331,0.039554774,0.179935287,0.037213237,0.521367185
+BDB,LMDTA - IDDTA (BD),0.76108371,0.003677725,0.250004744,0.022258109,0.122979933,0.020689633,0.164531214,0.023321004,0.517944237

supplementary_info/VocabSizeSearch.csv

@@ -0,0 +1,20 @@
+,,,Warm,,Cold Ligand,,Cold Protein,,Cold Both,
+Dataset,Chemical Vocab Size,Protein Vocab Size,CI,R2,CI,R2,CI,R2,CI,R2
+BDB,8000,8000,0.886 (0.009),0.767 (0.018),0.654 (0.081),-0.101 (0.206),0.726 (0.020),0.118 (0.107),0.557 (0.088),-0.299 (0.295)
+BDB,8000,16000,0.881 (0.006),0.754 (0.013),0.641 (0.092),0.005 (0.230),0.708 (0.030),0.040 (0.126),0.531 (0.049),-0.232 (0.288)
+BDB,8000,32000,0.883 (0.006),0.774 (0.013),0.657 (0.083),-0.143 (0.202),0.653 (0.060),-0.256 (0.411),0.522 (0.054),-0.442 (0.349)
+BDB,16000,8000,0.876 (0.007),0.762 (0.021),0.641 (0.116),-0.206 (0.376),0.696 (0.021),0.120 (0.100),0.520 (0.087),-0.318 (0.335)
+BDB,16000,16000,0.881 (0.011),0.769 (0.012),0.619 (0.070),-0.233 (0.160),0.702 (0.052),0.095 (0.261),0.502 (0.057),-0.505 (0.277)
+BDB,16000,32000,0.881 (0.004),0.766 (0.017),0.613 (0.097),-0.219 (0.224),0.673 (0.042),-0.224 (0.207),0.524 (0.081),-0.502 (0.299)
+BDB,32000,8000,0.887 (0.007),0.762 (0.018),0.642 (0.088),-0.116 (0.155),0.707 (0.029),0.072 (0.172),0.518 (0.064),-0.221 (0.104)
+BDB,32000,16000,0.886 (0.007),0.760 (0.017),0.643 (0.066),-0.029 (0.103),0.704 (0.027),0.042 (0.139),0.516 (0.054),-0.177 (0.133)
+BDB,32000,32000,0.885 (0.002),0.772 (0.015),0.666 (0.043),-0.150 (0.131),0.671 (0.052),-0.231 (0.488),0.512 (0.052),-0.396 (0.162)
+KIBA,8000,8000,0.880 (0.008),0.761 (0.017),0.733 (0.026),0.256 (0.110),0.690 (0.034),0.229 (0.100),0.613 (0.031),-0.015 (0.102)
+KIBA,8000,16000,0.881 (0.004),0.760 (0.017),0.735 (0.018),0.268 (0.094),0.692 (0.025),0.209 (0.087),0.623 (0.033),0.005 (0.150)
+KIBA,8000,32000,0.881 (0.005),0.760 (0.016),0.735 (0.025),0.274 (0.105),0.680 (0.020),0.185 (0.077),0.605 (0.033),-0.006 (0.117)
+KIBA,16000,8000,0.882 (0.007),0.767 (0.019),0.739 (0.018),0.275 (0.104),0.699 (0.029),0.263 (0.071),0.613 (0.022),0.050 (0.104)
+KIBA,16000,16000,0.877 (0.003),0.762 (0.010),0.736 (0.017),0.294 (0.065),0.687 (0.025),0.202 (0.079),0.618 (0.030),0.009 (0.111)
+KIBA,16000,32000,0.879 (0.006),0.760 (0.017),0.737 (0.016),0.297 (0.077),0.686 (0.031),0.194 (0.078),0.615 (0.036),0.016 (0.110)
+KIBA,32000,8000,0.880 (0.004),0.759 (0.011),0.740 (0.018),0.298 (0.088),0.702 (0.022),0.254 (0.033),0.626 (0.032),0.038 (0.126)
+KIBA,32000,16000,0.880 (0.005),0.762 (0.017),0.739 (0.013),0.297 (0.069),0.696 (0.035),0.219 (0.114),0.604 (0.055),-0.004 (0.145)
+KIBA,32000,32000,0.878 (0.003),0.760 (0.017),0.736 (0.015),0.261 (0.089),0.683 (0.029),0.198 (0.067),0.604 (0.031),-0.042 (0.054)