An interpretable machine learning method for homo-trimeric protein interface residue-residue interaction prediction.

Protein-protein interaction plays an important role in life activities. A more fine-grained analysis, such as residues and atoms level, will better benefit us to understand the mechanism for inter-protein interaction and drug design. The development of efficient computational methods to reduce trials and errors, as well as assisting experimental researchers to determine the complex structure are some of the ongoing studies in the field. The research of trimer protein interface, especially homotrimer, has been rarely studied. In this paper, we proposed an interpretable machine learning method for homo-trimeric protein interface residue pairs prediction. The structure, sequence, and physicochemical information are intergraded as feature input fed to model for training. Graph model is utilized to present spatial information for intra-protein. Matrix factorization captures the different features' interactions. Kernel function is designed to auto-acquire the adjacent information of our target residue pairs. The accuracy rate achieves 54.5% in an independent test set. Sequence and structure alignment exhibit the ability of model self-study. Our model indicates the biological significance between sequence and structure, and could be auxiliary for reducing trials and errors in the fields of protein complex determination and protein-protein docking, etc. SIGNIFICANCE: Protein complex structures are significant for understanding protein function and promising functional protein design. With data increasing, some computational tools have been developed for protein complex residue contact prediction, which is one of the most significant steps for complex structure prediction. But for homo-trimeric protein, the sequence-based deep learning predictors are infeasible for homologous sequences, and the algorithm black box prevents us from understanding of each step operation. In this way, we propose an interpreting machine learning method for homo-trimeric protein interface residue-residue interaction prediction, and the predictor shows a good performance. Our work provides a computational auxiliary way for determining the homo-trimeric proteins interface residue pairs which will be further verified by wet experiments, and and gives a hand for the downstream works, such as protein-protein docking, protein complex structure prediction and drug design.

Psychological and Functional Comparison between Minimally Invasive and Open Transforaminal Lumbar Interbody Fusion for Single-Level Lumbar Spinal Stenosis.

OBJECTIVE: The aim of this study was to investigate whether treatment with minimally invasive transforaminal lumbar interbody fusion (Mis-TLIF) causes patients suffering from lumbar spinal stenosis (LSS) to experience less anxiety and better clinical efficacy than open transforaminal lumbar interbody fusion (TLIF). METHODS: In this retrospective cohort study, we analyzed 86 patients, including 46 male patients and 41 female patients, who suffered from single-segmental lumbar spinal stenosis in our department between January 2016 and January 2018. They were divided into two groups: a control group (n = 46), for patients who underwent open TLIF surgery, and an experimental group (n = 40), for patients who underwent Mis-TLIF surgery. All patients were evaluated based on operation time, intraoperative blood loss, hospital stay, visual analogue scale (VAS), Oswestry disability index (ODI), hospital anxiety depression scale (HADS), fusion rate, and complications (screw misplacement and loosening, cerebrospinal fluid leakage, infection, and delayed wound healing). Patient characteristics were compared within and between groups. RESULTS: The average incision length was 3.64 ± 0.476 cm in the experimental group, which was smaller than that (8.11 ± 2.406 cm) in the control group (P < 0.05). The operation time of the experimental group was a little longer than that of the control group. The intraoperative blood loss and hospital stay in the experimental group were less than those in the control group. The mean preoperative low back pain VAS score was 7.525 ± 1.432 in the experimental group and 7.087 ± 1.799 in the control group (P > 0.05). The low back pain VAS scores on postoperative day 3 and at 3, 6, and 12 months postoperatively were 5.000 ± 0.987, 4.075 ± 0.997, 2.150 ± 0.834, and 1.450 ± 0.639 in the experimental group, respectively; these scores were lower than those in the control group (6.870 ± 1.572, P < 0.05; 4.630 ± 1.103, P < 0.05; 2.630 ± 1.103, P < 0.05; and 2.326 ± 1.034, P < 0.05, respectively). There was no obvious difference in the leg pain VAS scores between the two groups at all follow-up points. The mean preoperative ODI score was 58.700% ± 19.703% in the experimental group and 61.696% ± 17.583% in the control group (P > 0.05). The ODI scores at postoperative months 3, 6, and 12 were 25.225% ± 5.554%, 20.150% ± 7.698%, and 16.125% ± 9.565% in the experimental group; these scores were lower than those in the control group (49.130% ± 14.805%, P < 0.05; 34.044% ± 15.148%, P < 0.05; and 29.282% ± 132.567%, P < 0.05, respectively). The mean preoperative HADS score was 14.475 ± 3.113 in the experimental group and 13.391 ± 2.824 in the control group (P > 0.05). However, the mean HADS scores on postoperative day 3 in the experimental group was 8.500 ± 2.000, decreasing obviously compared to the preoperative scores (P < 0.05). The mean postoperative HADS score on postoperative day 3 in the control group was 12.734 ± 1.949, which had not decreased significantly compared to the preoperative score (P > 0.05). The HADS scores in the experimental group was lower than that in the control group on postoperative day 3 (P < 0.05). In the correlation analysis, the incision length was correlated to the HADS scores on postoperative day 3 (r = 0.527, P < 0.05). The HADS scores on postoperative day 3 were positively correlated with the low back pain VAS scores on the same day (r = 0.388, P < 0.05). The HADS scores on postoperative day 3were positively correlated with the ODI scores at 3-month (r = 0.460, P < 0.05), 6-month (r = 0.429, P < 0.05), and 12-month follow up (r = 0.349, P < 0.05). Fusion rates were not significantly different between the two groups. There was no screw misplacement and loosening, infection, or delayed wound healing in either group. The cerebrospinal fluid leakage rate in the control group was higher than that in the experimental group. CONCLUSION: Patients undergoing Mis-TLIF experience less anxiety and have better outcomes than those who undergo open TLIF. The lower level of anxiety experienced by patients undergoing Mis-TLIF is positively correlated with postoperative VAS and ODI scores.

Recurrent solitary fibrous tumor of the spinal cord: A case report and literature review.

Solitary fibrous tumor (SFT) is a benign mesenchymal neoplasm occurring anywhere in the body, such as the visceral pleura, while it is extremely rare in the central nervous system, especially within the spinal cord. Here, we present a case of recurrent spinal SFT in a 44-year-old woman who had the tumor resected 5 years before. This time, her magnetic resonance imaging revealed an intradural tumor at the level of C6-7. A secondary resection was performed completely, and the patient's neurological conditions recovered fully after resection. Histological and immunohistochemical findings revealed an SFT. Although rare, the preferred treatment for recurrent tumor in SFT is surgery, and postoperative follow-up is necessary for early detection of tumor progression.

Long-term outcome of percutaneous balloon compression for trigeminal neuralgia patients elder than 80 years: A STROBE-compliant article.

This article evaluates the long-term outcome of percutaneous balloon compression (PBC) for trigeminal neuralgia (TN) patients elder than 80 years. A total of 138 elderly patients aged above 80 years with primary TN, who were admitted to Neurosurgery Department, Hangzhou First People's Hospital from January 2007 to December 2011 for PBC treatment, were retrospectively analyzed in this study. The postoperative cure rate of immediate pain was 98.6% (Barrow Neurological Institute [BNI] classes I, II); according to the follow-up, the pain cure rates at 1, 2, 3, 4, and 5 years after surgery were 93.5%, 90.4%, 84.7%, 80.4%, and 72.9%, respectively. In our group, postoperative diplopia was reported in 1 case, masticatory muscle weakness in 3 cases, and herpes labialis in 19 cases. A total of 100% of pain-cured patients exhibited facial numbness and facial hypoesthesia. No serious complications occurred in this group of patients. PBC is an effective and safe procedure for TN treatment and can be employed as the preferred regimen for elderly TN patients aged above 80 years in poorer physical condition.