Neuropeptides acting as signaling molecules in the nervous system of various animals play crucial roles in a wide range of physiological functions and hormone regulation behaviors. Neuropeptides offer many opportunities for the discovery of new drugs and targets for the treatment of neurological diseases. In recent years, there have been several data-driven computational predictors of various types of bioactive peptides, but the relevant work about neuropeptides is little at present. In this work, we developed an interpretable stacking model, named NeuroPpred-Fuse, for the prediction of neuropeptides through fusing a variety of sequence-derived features and feature selection methods. Specifically, we used six types of sequence-derived features to encode the peptide sequences and then combined them. In the first layer, we ensembled three base classifiers and four feature selection algorithms, which select non-redundant important features complementarily. In the second layer, the output of the first layer was merged and fed into logistic regression (LR) classifier to train the model. Moreover, we analyzed the selected features and explained the feasibility of the selected features. Experimental results show that our model achieved 90.6% accuracy and 95.8% AUC on the independent test set, outperforming the state-of-the-art models. In addition, we exhibited the distribution of selected features by these tree models and compared the results on the training set to that on the test set. These results fully showed that our model has a certain generalization ability. Therefore, we expect that our model would provide important advances in the discovery of neuropeptides as new drugs for the treatment of neurological diseases.
Candida parapsilosis carbonyl reductases (CpRCR) have been widely used for the reductive conversion of ketone precursors and chiral alcohol products in pharmaceutical industries. The enzymatic enantioselectivity is believed to be related to the shape complementation between the cavities in the enzymes and the substitutions of the ketone substrates. In this work, we reported an unexpected enantioselectivity in the enzyme reductions of dihydrofuran-3(2H)-one (DHF) to (S)-tetrahydrofuran-3-ol (DHF-ol, enantiomeric excess: 96.4%), while dihydrothiophen-3(2H)-one substrate (DHT) was unproductive under the same experimental conditions. To rationalize the exclusive S-configuration and the specific reactivity of DHF, we carried out molecular dynamics simulations for the reacting complexations of DHF with CpRCR, and DHT with CpRCR. Our calculations indicate that DHF preferentially binds to the small cavity near L119, F285, and W286, while the large cavity near the α1 helix was mainly occupied by solvent water molecules. Moreover, the pre-reaction state analysis suggests that the pro-S conformations were more abundant than the pro-R, in particular for DHF. This suggests that the non-polar interaction of substrate C4-C5 methylene contacting the hydrophobic side-chains of L119-F285-W286, and the polar interaction of funanyl oxygen exposing the solvent environment play important roles in the enantioselectivity and reactivity. The phylogenetic tree of CpRCR homologues implies that a variety of amino acid combinations at positions 285 and 286 were available and thereby potentially useful for redesigning enantioselective reductions of 5-membered-ring heterocyclic ketones.
Here, we present a protocol to examine asymmetric pairwise pre-reaction and transition states in enzymatic catalysis. We describe steps to set up the calculated systems, run umbrella sampling molecular dynamics simulation, and conduct quantum mechanics/molecular mechanics calculations. We also provide analytical scripts to yield potential of mean force of pre-reaction states and reaction barriers. This protocol can generate quantum-mechanistic data for constructing pre-reaction state/transition state machine learning models. For complete details on the use and execution of this protocol, please refer to Luo et al. (2022).1.