Efficient ground-glass opacities (GGOs) diagnosis is challenging. A diagnostic method distinguishing malignant from benign GGOs is warranted. In this study, we sought to construct a noninvasive method based on tumor educated platelet (TEP) RNA profiles for malignant GGOs diagnosis and explore the molecular mechanism of the potential biomarker for the first time. Based on TEP RNA-sequencing (TEP RNA-seq) in benign and malignant GGOs, a classification model was constructed using differentially expressed genes (DEGs) and was used to evaluate diagnostic performance. High-throughput quantitative polymerase chain reaction (HT-qPCR) verified 23 genes selected from the top 60 DEGs between benign and malignant GGOs. The correlation between 17 verified DEGs and 22 key glycolytic genes was analyzed. Tripartite motif-containing 27 (TRIM27) overexpressing and knockdown (KD) cell models were constructed using A549 and PC-9 cells, respectively in which cell growth, apoptosis, migration and invasion were evaluated. The protein levels of HK-1/2, PKM1/2, LDHA and GLUT1 were evaluated by western blot. Glycolysis was evaluated through adenosine triphosphate (ATP), reactive oxygen species (ROS), lactate acid (LD) production, glucose uptake, and lactate dehydrogenase (LDH) activity assays. RNA-seq was performed in loss-of TRIM27-KD PC-9 cells to clarify the downstream factors of TRIM27 which was verified using western blot and immunofluorescence double staining. In 81 samples, the 1,647-DEG-based classification model exhibited area under the curve (AUC), sensitivity, and specificity values of 0.99 [95% confidence interval (CI): 0.972-1.000], 100%, and 91%, respectively, while the top 60-DEG-based classification model exhibited AUC, sensitivity, and specificity values of 0.986 (95% CI: 0.962-1.000), 98%, and 91%, respectively. TRIM27 achieved AUC of 0.87 in the diagnosis of malignant GGOs, with 83.93% sensitivity, 78.79% specificity, 81.15% accuracy, 77.05% positive predictive value (PPV) and 85.25% negative predictive value (NPV). TRIM27 was highly expressed in non-small cell lung cancer (NSCLC) cells, and accelerated cell migration and invasion. In addition, TRIM27 was found to promote glycolysis in NSCLC cells partially through HMOX1 which was negatively correlated with TRIM27. We constructed a novel TEP RNA-seq based classifier for malignant GGOs diagnosis. TRIM27, an important target discovered, could accelerate migration, invasion and regulate glycolysis partially through HMOX1 in NSCLC cells, thus providing scientific support for TRIM27 as a diagnostic biomarker for malignant GGO diagnosis.