PITHIA: Protein Interaction Site Prediction Using Multiple Sequence Alignments and Attention

Abstract

Cellular functions are governed by proteins, and, while some proteins work independently,most work by interacting with other proteins. As a result it is crucially important to know theinteraction sites that facilitate the interactions between the proteins. Since the experimental methodsare costly and time consuming, it is essential to develop effective computational methods. We presentPITHIA, a sequence-based deep learning model for protein interaction site prediction that exploits thecombination of multiple sequence alignments and learning attention. We demonstrate that our newmodel clearly outperforms the state-of-the-art models on a wide range of metrics. In order to providemeaningful comparison, we update existing test datasets with new information regarding interactionsite, as well as

State‐of‐the‐art computational methods to predict protein–protein interactions with high accuracy and coverage

“…The trend in the field is that the current models the field is publishing involve attention. PITHIA demonstrated that the use of the transformer architecture with attention yielded the best results when testing different architectures of the model [67]. Therefore, for the foreseeable future, in order for models to compete with the current state-of-the-art, they will need to incorporate an attention mechanism.…”

“…Surprisingly, however, when testing the impact of additional features such as a PSSM, physiochemical characteristics, or evolutionary conservation, the use of MSAs alone yielded the best results [67]. PITHIA demonstrates the power of MSAs within the attention mechanism compared against other architectures and features.…”

“…The next strategy for protein sequence representation is the use of embeddings and encoders. Encoders such as one hot encodings can be used to represent protein sequences, while embeddings such as Word2Vec, FastText, and BERT [7,34,43,44,48,67,88] are commonly used to transform the protein sequence into an interpretable feature for a neural network. The main advantage to this strategy is that it keeps some resemblance of the entire protein sequence while also providing an interpretable feature representation for a neural network [43].…”

“…However, models that predict local interaction tend to use a combination of statistical, physiochemical representations as well as some representation of the overall protein sequence that captures local features of the protein (e.g., overall fold or domains). As described in the previous sections, protein sequence representations encompass encoding methods such as metric representations, text embeddings, and neural network feature embeddings, but some groups have also leveraged raw protein sequences [39,42,44,67,86,87]. Using unprocessed protein sequences for PPI prediction creates an issue for neural network architectures since most models depend on an input of fixed length.…”

“…Case example of MSAs: PITHIA An example of a neural network architecture incorporating MSAs is PITHIA [67], which uses not only MSAs but also uses other side features such as embeddings from protein language models and other representations such as a PSSM. While PITHIA does not exactly make PPI predictions, it can find protein interaction sites on a singular protein.…”