As a newly-identified protein post-translational modification, malonylation is involved in a variety of biological functions. Recognizing malonylation sites in substrates represents an initial but crucial step in elucidating the molecular mechanisms underlying protein malonylation. In this study, we constructed a deep learning (DL) network classifier based on long short-term memory (LSTM) with word embedding (LSTMWE) for the prediction of mammalian malonylation sites. LSTMWE performs better than traditional classifiers developed with common pre-defined feature encodings or a DL classifier based on LSTM with a one-hot vector. The performance of LSTMWE is sensitive to the size of the training set, but this limitation can be overcome by integration with a traditional machine learning (ML) classifier. Accordingly, an integrated approach called LEMP was developed, which includes LSTMWE and the random forest classifier with a novel encoding of enhanced amino acid content. LEMP performs not only better than the individual classifiers but also superior to the currently-available malonylation predictors. Additionally, it demonstrates a promising performance with a low false positive rate, which is highly useful in the prediction application. Overall, LEMP is a useful tool for easily identifying malonylation sites with high confidence. LEMP is available at http://www.bioinfogo.org/lemp.

Early region 1A (E1A) is the first viral protein produced upon human adenovirus (HAdV) infection. This multifunctional protein transcriptionally activates other HAdV early genes and reprograms gene expression in host cells to support productive infection. E1A functions by interacting with key cellular regulatory proteins through short linear motifs (SLiMs). In this study, the molecular determinants of interaction between E1A and BS69, a cellular repressor that negatively regulates E1A transactivation, were systematically defined by mutagenesis experiments. We found that a minimal sequence comprised of MPNLVPEV, which contains a conserved PXLXP motif and spans residues 112–119 in HAdV-C5 E1A, was necessary and sufficient in binding to the myeloid, Nervy, and DEAF-1 (MYND) domain of BS69. Our study also identified residues P113 and L115 as critical for this interaction. Furthermore, the HAdV-C5 and -A12 E1A proteins from species C and A bound BS69, but those of HAdV-B3, -E4, -D9, -F40, and -G52 from species B, E, D, F, and G, respectively, did not. In addition, BS69 functioned as a repressor of E1A-mediated transactivation, but only for HAdV-C5 and HAdV-A12 E1A. Thus, the PXLXP motif present in a subset of HAdV E1A proteins confers interaction with BS69, which serves as a negative regulator of E1A mediated transcriptional activation.