In this review, the concept and idea of chemical modeling and model population analysis ( MPA) were introduced, and the recent applications of MPA-based methods to different aspects of chemical modeling were listed, including outlier detection, variable selection, model evaluation and applicability domain. In addition, the feasibility and applicability of MPA to different kinds of dataset was illustrated, such as near infrared spectroscopy, quantitative structure-activity relationship and metabolomics, which provided a better idea and framework to develop a new algorithm in chemical modeling.

BACKGROUND: miRNA-136-5 p plays a crucial regulatory role in pathological changes, inflammatory response and regeneration after spinal cord injury. OBJECTIVE: To investigate the effect of miRNA-136-5 p on the expression of cytokines in serum and NF-κB protein in spinal cord in rats with spinal cord injury and to explore the molecular mechanism. METHODS: Thirty-six male Sprague-Dawley rats, of SPF grade were provided by Laboratory Animal Center of Guangxi Medical University. The lentiviral vector system was prepared and transfected into spinal cord injured rats. Thirty-six rat models of spinal cord injury were established by modified Allen's method. Basso Beattie Bresnahan scores were performed. Rats were randomly divided into normal control, modeling (LV-ctrl plus spinal cord injury), overexpression (spinal cord injury plus LV-miRNA-136-5 p), and inhibition (spinal cord injury plus LV-sponge) groups (n=9/group). Seven days before surgery and the day of surgery, the overexpression and inhibition groups were continuously injected with the lentivirus suspension into the injured area, and the normal control and modeling groups were injected with the same amount of normal saline. Three rats were sacrificed at 1, 3 and 7 days, and blood and spinal cord tissues were taken. The levels of interleukin-1β, interleukion-6 and interferon-α in rat serum were determined by ELISA. The expression of NF-κB protein was detected by western blot assay and double immunofluorescence. RESULTS AND CONCLUSION: (1) There was no significant difference in preoperative Basso Beattie Bresnahan scores (P＞ 0.05). In the modeling group, the rats showed prone walking, vary degrees of urinary retention, and spinal shock, with complete loss of function of both hind limbs and muscle strength of 0. (2) Compared with the normal control group, the levels of inflammatory factors in the other groups were increased significantly (P ＜ 0.05). The expression levels of inflammatory factors were highest in the overexpression group, followed by modeling group, and lowest in the inhibition group. (3) Results of western blot assay and double immunofluorescence showed that the expression level of NF-κB protein in the modeling, overexpression and inhibition groups was significantly higher than that in the normal control group (P ＜ 0.05), and the level was highest in the overexpression group. (4) In summary, miRNA-136-5 p can affect inflammatory factors and NF-κB in rats with acute spinal cord injury.