Objective To explore the value of drug susceptibility test (DST) by micropore-plate method and whole-genome sequencing (WGS) in detecting the drug resistance of anti-tuberculosis drugs in children. Methods Sixty-one partially preserved strains with positive culture of Mycobacterium tuberculosis (MTB) were collected. The drug resistance of isoniazid, rifampicin, ethambutol, and streptomycin were detected by proportional method, micropore-plate method DST and WGS. The variants of four drug resistance related genes in WGS were summarized, and the clinical data, treatment plan and recovery of the affected children were collected. The reasons for the differences of drug resistance detected by the three methods were analyzed. Results Among 61 MTB strains, 78.69% (n=48) were pan-susceptible strains, 1.64% (n=1) was isoniazid-resistant strain, 1.64% (n=1) was streptomycin resistant strain, 6.56% (n=4) were isoniazid and streptomycin resistant strains, 3.28% (n=2) were isoniazid, rifampicin, and ethambutol resistant strains, 3.28% (n=2) were isoniazid, rifampicin and streptomycin resistant strains, and 4.92% (n=3) were resistant to the four drugs. The sensitivity, specificity, positive predictive value, negative predictive value, and the Kappa value of micropore-plate method for predicting isoniazid resistance were 100%, 92.0%, 73.3%, 100% and 0.81, for rifampicin were 85.7%, 98.2%, 85.7%, 98.2% and 0.84, for ethambutol were 0.0%, 100%, 0.0%, 91.8% and 0.0, for streptomycin were 80.0%, 98.0%, 88.9%, 96.2% and 0.81, respectively. The sensitivity, specificity, positive predictive value, negative predictive value, and the Kappa value of WGS for predicting isoniazid resistance were 90.9%, 94.0%, 76.9%, 97.9% and 0.79, for rifampicin were 100%, 98.2%, 87.5%, 100% and 0.92, for ethambutol were 60.0%, 94.6%, 50.0%, 96.4% and 0.50, for streptomycin were 80.0%, 98.0%, 88.9%, 96.2% and 0.81, respectively. Conclusion Highly consistent with ratio method, micropore-plate method and WGS have higher sensitivity to predict isoniazid, rifampin, and streptomycin resistant tuberculosis in children, which can replace traditional proportional method in clinical practice. However, it is suggested to use proportional method combined with WGS to comprehensively evaluate the ethambutol resistance of MTB, as the consistency of micropore-plate and ratio methods is poor.