![](https://csdnimg.cn/release/download_crawler_static/12258507/bg4.jpg)
4
1.
pK
d
= −log10(
K
d
1e9
) (1)
Figure 1A (left panel) illustrates the distribution of the binding affinity values in pK
d
form.
The peak at pK
d
value 5 (10000nM) constitutes more than half of the data set (20931 out
of 30056). These values correspond to the negative pairs that either have very weak binding
affinities (K
d
> 10000nM) or are not observed in the primary screen [47]. As such they are
true negatives.
The distribution of the KIBA scores is depicted in the right panel of Figure 1A. [29]
pre-processed the KIBA scores as f ollows: (i) for each KIBA score, its negative was taken,
(ii) the m inimum value among the negatives was chosen, and (iii) the absolute value of the
minimum was added to all negative scores, thus constructing the final form of the KIBA
scores.
The compound S MIL ES strings of the Davis data set were extracted from the Pubchem
compound database b ased on their Pu bchem CIDs [5]. For KIBA, first the CHEMBL IDs
were converted into Pubchem CIDs and then, the corresponding CIDs were used to extract
the SMILES strings. Figure 1B illustrates the distribution of the lengths of the SMILES
strings of the compounds in the Davis (left) and KIBA (right) data sets. For the compounds
of the Davis dataset, the maximum length of a SMILES is 103, while the average length is
equal to 64. For the comp ou nds of KIBA, the maximum length of a SMILES is 590, while
the average length is equal to 58.
The p rotein sequences of the Davis data set wer e extracted from the UniProt protein
database based on gene names/RefSeq accession numbers [2]. Similarly, the UniProt IDs
of the targets in the KIBA data set were used to collect the protein sequences. Figure 1C
left pan el shows the lengths of the sequences of the proteins in the Davis data set. The
maximum length of a protein sequence is 2549 and the average length is 788 characters.
Figure 1C right panel depicts the distribution of protein sequence length in KIBA targets.
The maximum length of a protein sequence is 4128 and the average length is 728 characters.
We should also note that the Smith-Waterman (S-W) similarity among proteins of the
KIBA data set is at m ost 60% for 99% of the protein pairs. The target similarity is at most
60% for 92% of the protein pairs for the Davis data set. These statistics indicate that both
data sets are non-redundant.
Input Representation
We used integer/label encoding that uses integers for the categories to represent inputs. We
scanned approximately 2M SMILES sequences th at we collected from Pubchem and com-
piled 64 labels (unique letters). For protein sequences, we scanned 550K protein sequences
from UniProt and extracted 25 categories (unique letters).
Here we represent each label with a corresponding integer (e.g. “C”:1, “H”:2, ‘N”:3 etc.).
The label encoding for the example S MIL ES, “CN=C=O”, is given below.
C N = C = O
=
1 3 63 1 63 5
Protein sequences are encoded in a sim ilar way using label encodings. Both SMILES and
protein sequences have varying lengths. Hence, in order to create an effective representation