Matchms 工作流

# Matchms 常用工作流 本文档提供了使用 matchms 进行质谱分析的常用工作流详细示例。 ## 工作流 1：基础谱图库匹配 将未知谱图与参考库进行匹配以鉴定化合物。 python from matchms.importing import load_from_mgf from matchms.filtering import default_filters, normalize_intensities from matchms.filtering import select_by_relative_intensity, require_minimum_number_of_peaks from matchms import calculate_scores from matchms.similarity import CosineGreedy # 加载参考库 print("正在加载参考库...") library = list(load_from_mgf("reference_library.mgf")) # 加载查询谱图（未知物） print("正在加载查询谱图...") queries = list(load_from_mgf("unknown_spectra.mgf")) # 处理库谱图 print("正在处理库...") processed_library = [] for spectrum in library: spectrum = default_filters(spectrum) spectrum = normalize_intensities(spectrum) spectrum = select_by_relative_intensity(spectrum, intensity_from=0.01) spectrum = require_minimum_number_of_peaks(spectrum, n_required=5) if spectrum is not None: processed_library.append(spectrum) # 处理查询谱图 print("正在处理查询...") processed_queries = [] for spectrum in queries: spectrum = default_filters(spectrum) spectrum = normalize_intensities(spectrum) spectrum = select_by_relative_intensity(spectrum, intensity_from=0.01) spectrum = require_minimum_number_of_peaks(spectrum, n_required=5) if spectrum is not None: processed_queries.append(spectrum) # 计算相似度 print("正在计算相似度...") scores = calculate_scores(references=processed_library, queries=processed_queries, similarity_function=CosineGreedy(tolerance=0.1)) # 获取每个查询的顶级匹配结果 print("n顶级匹配:") for i, query in enumerate(processed_queries): top_matches = scores.scores_by_query(query, sort=True)[:5] query_name = query.get("compound_name", f"Query {i}") print(f"n{query_name}:") for ref_idx, score in top_matches: ref_spectrum = processed_library[ref_idx] ref_name = ref_spectrum.get("compound_name", f"Ref {ref_idx}") print(f" {ref_name}: {score:.4f}") --- ## 工作流 2：质量控制与数据清洗 在分析前对谱图数据进行过滤和清洗。 python from matchms.importing import load_from_mgf from matchms.exporti