Predicting antimicrobial resistance in Pseudomonas aeruginosa with machine learning-enabled molecular diagnostics

Datasets for manuscript "Predicting antimicrobial resistance in Pseudomonas aeruginosa with machine learning-enabled molecular diagnostics" <strong>Metadata.zip</strong> <strong>phenotypes.txt: </strong>tabular file containing binary resistance phenotypes based on CLSI guidelines, where the rows are the isolates and the columns correspond to different drugs. Resistance : 1, susceptibility: 0, missing: intermediate resistant <strong>Features_gpa_exp_snps.zip </strong> We provide the processed molecular data as Numpy compressed files (npz.). You can use the Numpy load method to read in these tables https://docs.scipy.org/doc/numpy/reference/generated/numpy.load.htm. The row (strains_list) and column labels (feature_lists) are stored separately. <strong>genexp</strong>: gene expression table directory genexp_feature_vect.npz: The feature matrix in the numpy format genexp_feature_list.txt: The columns of the feature matrix (features) genexp_strains_list.txt: The rows of the feature matrix (isolates) <strong>gpa: </strong>gene presence/absence table directory gpa_feature_vect.npz: The feature matrix in the numpy format gpa_feature_list.txt: The columns of the feature matrix (features) gpa_strains_list.txt: The rows of the feature matrix (isolates) <strong>snps: </strong>SNPs table directory snps_feature_vect.npz: The feature matrix in the numpy format snps_feature_list.txt: The columns of the feature matrix (features) snps_strains_list.txt: The rows of the feature matrix (isolates)

对应论文《基于机器学习赋能分子诊断预测铜绿假单胞菌抗菌药物耐药性》的数据集

<strong>Metadata.zip</strong>
<strong>phenotypes.txt</strong>：基于临床实验室标准化协会（Clinical and Laboratory Standards Institute, CLSI）指南构建的二元耐药表型表格文件，行代表菌株分离株，列对应不同抗菌药物。其中耐药性标注为1，敏感性标注为0，缺失值代表中介耐药。
<strong>Features_gpa_exp_snps.zip</strong>：本压缩包包含处理完成的分子数据，格式为NumPy压缩文件（npz），可通过NumPy的load方法读取此类表格，读取方法详见https://docs.scipy.org/doc/numpy/reference/generated/numpy.load.htm。特征矩阵的行标签（对应strains_list文件）与列标签（对应feature_lists文件）分别单独存储。
<strong>genexp</strong>：基因表达表目录
- genexp_feature_vect.npz：NumPy格式的特征矩阵
- genexp_feature_list.txt：特征矩阵的列标签（即特征名称）
- genexp_strains_list.txt：特征矩阵的行标签（即菌株分离株）
<strong>gpa</strong>：基因存在/缺失表目录
- gpa_feature_vect.npz：NumPy格式的特征矩阵
- gpa_feature_list.txt：特征矩阵的列标签（即特征名称）
- gpa_strains_list.txt：特征矩阵的行标签（即菌株分离株）
<strong>snps</strong>：单核苷酸多态性（Single Nucleotide Polymorphisms, SNPs）表目录
- snps_feature_vect.npz：NumPy格式的特征矩阵
- snps_feature_list.txt：特征矩阵的列标签（即特征名称）
- snps_strains_list.txt：特征矩阵的行标签（即菌株分离株）