将大列表的元素取消嵌套到数据框中答案

【问题标题】：Unnesting elements of large list into dataframe将大列表的元素取消嵌套到数据框中
【发布时间】：2021-12-11 02:55:14
【问题描述】：

我目前正在处理从 crossref API 检索到的大型数据集，我在其中检索了基于 DOI 搜索的科学论文信息。

目前，大列表包含约 3500 个元素。这些元素中的每一个都是它们自己的列表，由元数据“元”、实际相关数据“数据”和不相关的列表“方面”组成。

这是基于两个 DOI 的两个列表的示例：

list(`10.1158/1055-9965.EPI-08-0303` = list(meta = NULL, data = structure(list(
    alternative.id = "10.1158/1055-9965.EPI-08-0303", container.title = "Cancer Epidemiology Biomarkers & Prevention", 
    created = "2008-11-06", deposited = "2020-12-24", published.print = "2008-11", 
    published.online = "2008-11-06", doi = "10.1158/1055-9965.epi-08-0303", 
    indexed = "2021-10-17", issn = "1055-9965,1538-7755", issue = "11", 
    issued = "2008-11", member = "1086", page = "3216-3223", 
    prefix = "10.1158", publisher = "American Association for Cancer Research (AACR)", 
    score = "1", source = "Crossref", reference.count = "31", 
    references.count = "31", is.referenced.by.count = "50", subject = "Oncology,Epidemiology", 
    title = "20 Years into the Gambia Hepatitis Intervention Study: Assessment of Initial Hypotheses and Prospects for Evaluation of Protective Effectiveness Against Liver Cancer", 
    type = "journal-article", url = "http://dx.doi.org/10.1158/1055-9965.epi-08-0303", 
    volume = "17", language = "en", short.container.title = "Cancer Epidemiol Biomarkers Prev", 
    author = list(structure(list(given = c("Simonetta", "Patrizia", 
    "Ebrima", "Andrew J.", "Gregory D.", "Maimuna", "Ruggero", 
    "Amelie", "Omar", "Marianne", "Hilton", "Pierre"), family = c("Viviani", 
    "Carrieri", "Bah", "Hall", "Kirk", "Mendy", "Montesano", 
    "Plymoth", "Sam", "Van der Sande", "Whittle", "Hainaut"), 
        sequence = c("first", "additional", "additional", "additional", 
        "additional", "additional", "additional", "additional", 
        "additional", "additional", "additional", "additional"
        )), class = c("tbl_df", "tbl", "data.frame"), row.names = c(NA, 
    -12L))), link = list(structure(list(URL = "https://syndication.highwire.org/content/doi/10.1158/1055-9965.EPI-08-0303", 
        content.type = "unspecified", content.version = "vor", 
        intended.application = "similarity-checking"), class = c("tbl_df", 
    "tbl", "data.frame"), row.names = c(NA, -1L)))), class = c("tbl_df", 
"tbl", "data.frame"), row.names = c(NA, -1L)), facets = NULL), 
    `10.1016/j.canlet.2007.10.044` = list(meta = NULL, data = structure(list(
        alternative.id = "S0304383507005253", container.title = "Cancer Letters", 
        created = "2008-01-14", deposited = "2019-01-01", published.print = "2008-03", 
        doi = "10.1016/j.canlet.2007.10.044", indexed = "2021-10-07", 
        issn = "0304-3835", issue = "1", issued = "2008-03", 
        member = "78", page = "21-25", prefix = "10.1016", publisher = "Elsevier BV", 
        score = "1", source = "Crossref", reference.count = "20", 
        references.count = "20", is.referenced.by.count = "71", 
        subject = "Cancer Research,Oncology", title = "Detection of R337H, a germline TP53 mutation predisposing to multiple cancers, in asymptomatic women participating in a breast cancer screening program in Southern Brazil", 
        type = "journal-article", url = "http://dx.doi.org/10.1016/j.canlet.2007.10.044", 
        volume = "261", language = "en", short.container.title = "Cancer Letters", 
        author = list(structure(list(given = c("Edenir Inêz", 
        "Lavínia", "Maira", "Maria Isabel Waddington", "Magali", 
        "Ghyslaine", "Virginie", "Ernestina", "Juliana", "Ingrid Petroni", 
        "Roberto", "Pierre", "Patricia"), family = c("Palmero", 
        "Schüler-Faccini", "Caleffi", "Achatz", "Olivier", "Martel-Planche", 
        "Marcel", "Aguiar", "Giacomazzi", "Ewald", "Giugliani", 
        "Hainaut", "Ashton-Prolla"), sequence = c("first", "additional", 
        "additional", "additional", "additional", "additional", 
        "additional", "additional", "additional", "additional", 
        "additional", "additional", "additional")), class = c("tbl_df", 
        "tbl", "data.frame"), row.names = c(NA, -13L))), link = list(
            structure(list(URL = c("https://api.elsevier.com/content/article/PII:S0304383507005253?httpAccept=text/xml", 
            "https://api.elsevier.com/content/article/PII:S0304383507005253?httpAccept=text/plain"
            ), content.type = c("text/xml", "text/plain"), content.version = c("vor", 
            "vor"), intended.application = c("text-mining", "text-mining"
            )), class = c("tbl_df", "tbl", "data.frame"), row.names = c(NA, 
            -2L))), license = list(structure(list(date = "2008-03-01", 
            content.version = "tdm", delay.in.days = 0L, URL = "https://www.elsevier.com/tdm/userlicense/1.0/"), class = c("tbl_df", 
        "tbl", "data.frame"), row.names = c(NA, -1L))), reference = list(
            structure(list(key = c("10.1016/j.canlet.2007.10.044_bib1", 
            "10.1016/j.canlet.2007.10.044_bib2", "10.1016/j.canlet.2007.10.044_bib3", 
            "10.1016/j.canlet.2007.10.044_bib4", "10.1016/j.canlet.2007.10.044_bib5", 
            "10.1016/j.canlet.2007.10.044_bib6", "10.1016/j.canlet.2007.10.044_bib7", 
            "10.1016/j.canlet.2007.10.044_bib8", "10.1016/j.canlet.2007.10.044_bib9", 
            "10.1016/j.canlet.2007.10.044_bib10", "10.1016/j.canlet.2007.10.044_bib11", 
            "10.1016/j.canlet.2007.10.044_bib12", "10.1016/j.canlet.2007.10.044_bib13", 
            "10.1016/j.canlet.2007.10.044_bib14", "10.1016/j.canlet.2007.10.044_bib15", 
            "10.1016/j.canlet.2007.10.044_bib16", "10.1016/j.canlet.2007.10.044_bib17", 
            "10.1016/j.canlet.2007.10.044_bib18", "10.1016/j.canlet.2007.10.044_bib19", 
            "10.1016/j.canlet.2007.10.044_bib20"), doi.asserted.by = c("crossref", 
            "crossref", NA, NA, NA, "crossref", NA, NA, "crossref", 
            "crossref", "crossref", NA, NA, NA, "crossref", "crossref", 
            "crossref", "crossref", "crossref", NA), first.page = c("96", 
            "266", "1298", "877s", "1932", "12", "10", "608", 
            "2658", "1954", "133", "1703", "1365", "5358", "1215", 
            "607", "647", "607", "9330", "1213"), DOI = c("10.1016/j.canlet.2005.12.039", 
            "10.1159/000154228", NA, NA, NA, "10.1038/nsb730", 
            NA, NA, "10.1002/1097-0142(19901215)66:12<2658::AID-CNCR2820661232>3.0.CO;2-C", 
            "10.1038/sj.onc.1207305", "10.1016/0165-4608(93)90166-J", 
            NA, NA, NA, "10.1093/nar/16.3.1215", "10.1002/humu.10081", 
            "10.1590/S0004-27302004000500009", "10.1007/BF00202835", 
            "10.1073/pnas.161479898", NA), article.title = c("The TP53 mutation, R337H, is associated with Li–Fraumeni and Li–Fraumeni-like syndromes in Brazilian families", 
            "Is p53 polymorphism maintained by natural selection?", 
            "Prevalence and diversity of constitutional mutations in the p53 gene among 21 Li–Fraumeni families", 
            "Breast cancer screening in 10,000 women of an underserved population in South Brazil: The NMAMAPOA cohort", 
            "P53 germline mutations in childhood cancers and cancer risk for carrier individuals", 
            "A novel mechanism of tumorigenesis involving pH-dependent destabilization of a mutant p53 tetramer", 
            "Germline mutations in the TP53 gene", "Germ-line p53 mutations in 15 families with Li–Fraumeni syndrome", 
            "Choroid plexus tumors in the breast cancer-sarcoma syndrome", 
            "A TP53 polymorphism is associated with increased risk of colorectal cancer and with reduced levels of TP53 mRNA", 
            "Wilms’ tumor in the Li–Fraumeni cancer family syndrome", 
            "Simple sequence repeat polymorphism within the p53 gene", 
            "Rhabdomyosarcoma in children: epidemiologic study and identification of a familial cancer syndrome", 
            "A cancer family syndrome in twenty-four kindreds", 
            "A simple salting out procedure for extracting DNA from human nucleated cells", 
            "The IARC TP53 database: new online mutation analysis and recommendations to users", 
            "Founder effect for the highly prevalent R337H mutation of tumor suppressor p53 in Brazilian patients with adrenocortical tumors", 
            "Identification of a polymorphism in intron 2 of the p53 gene", 
            "An inherited p53 mutation that contributes in a tissue-specific manner to pediatric adrenal cortical carcinoma", 
            "Cancer in survivors of childhood soft tissue sarcoma and their relatives"
            ), volume = c("245", "44", "54", "23", "82", "9", 
            "25", "56", "66", "23", "67", "8", "43", "48", "16", 
            "19", "48", "93", "98", "79"), author = c("Achatz", 
            "Beckman", "Birch", "Caleffi", "Chompret", "DiGiammarino", 
            "Eeles", "Frebourg", "Garber", "Gemignani", "Hartley", 
            "Lazar", "Li", "Li", "Miller", "Olivier", "Pinto", 
            "Pleasants", "Ribeiro", "Strong"), year = c("2007", 
            "1994", "1994", "2005", "2000", "2002", "1995", "1995", 
            "1990", "2004", "1993", "1993", "1969", "1988", "1988", 
            "2002", "2004", "1994", "2001", "1987"), journal.title = c("Cancer Lett.", 
            "Hum. Hered.", "Cancer Res.", "J. Clin. Oncol.", 
            "Br. J. Cancer", "Nat. Struct. Biol.", "Cancer Surv.", 
            "Am. J. Hum. Genet.", "Cancer", "Oncogene", "Cancer Genet. Cytogenet.", 
            "Oncogene", "J. Natl. Cancer Inst.", "Cancer Res.", 
            "Nucleic Acids Res.", "Hum. Mutat.", "Arq. Bras. Endocrinol. Metabol.", 
            "Hum. Genet.", "Proc. Natl. Acad. Sci. USA", "J. Natl. Cancer Inst."
            ), issue = c(NA, "5", "5", "16 S", "12", "1", NA, 
            "3", "12", "10", "2", "6", "6", "18", "3", "6", "5", 
            "5", "16", "6")), class = c("tbl_df", "tbl", "data.frame"
            ), row.names = c(NA, -20L)))), class = c("tbl_df", 
    "tbl", "data.frame"), row.names = c(NA, -1L)), facets = NULL))

“数据”下的所有数据都与我相关，我想取消嵌套这些数据并构建一个大型数据框，其中 DOI 在一列中，“数据”下的数据在其他列中。

我尝试使用unnest 代码unnest(data)，但结果为Error in UseMethod("unnest") : no applicable method for 'unnest' applied to an object of class "list"

有没有简单的方法可以做到这一点？

【问题讨论】：

标签： r list unnest

【解决方案1】：

像这样？注意 - 最好包含一个包含玩具数据集的Minimal reprex，而不是您拥有的快照。这样问题可能会更快得到答案。

ll <- list(`10.1016/j.ejca.2017.11.029` = list(metadata = NULL,
     data = tibble(one = 1, two = 2)),
     `10.1016/j.ejca.2017.12.500` = list(metadata = NULL,
                 data = tibble(one = 3, two = 4)))
nms <- names(ll)
vals <- lapply(ll, `[[`, 2)

tibble::tibble(DOI = nms,
       data = vals)

# or shorten to
tibble::tibble(DOI = names(ll),
               data = lapply(ll, `[[`, 2))

# A tibble: 2 x 2
  DOI                        data            
  <chr>                      <named list>    
1 10.1016/j.ejca.2017.11.029 <tibble [1 x 2]>
2 10.1016/j.ejca.2017.12.500 <tibble [1 x 2]>

解释 > 这里的名称包含信息。相关信息是第二个列表元素。 lapply(ll, [[, 2) 等价于c(ll[[1]][[2]], ll[[2]][[2]], ... )

【讨论】：

谢谢！我已将一个玩具数据集添加到我根据您使用的玩具数据集改编的原始问题中。您的代码似乎正在处理这个玩具数据集，但是当我尝试将其应用于我的实际数据集时，我收到错误消息“In tibble:tibble(DOI = names(ll), data = lapply(l, : numeric expression has 2 个元素：只使用第一个元素。”我在这里遗漏了什么？难道是“数据”下的小标题每个都有不同的列数？
你在使用tibble::tibble(DOI = names(ll), data = lapply(ll, [[, 2))时得到了那个？在这种情况下，玩具示例与您的实际数据的结构不同，需要一个 dput 示例来查看函数需要如何调整。注意这里的 [[ 应该用反引号括起来，这是本网站上的一种特殊格式，因此无法正确打印。
我不知道dput 函数的存在。谢谢，这对创建 Minimal reprex 非常有帮助。我在原始帖子中添加了两个 DOI 的示例。
当我在您添加到原始帖子中的数据上运行代码时，它运行良好。 data 中的列数差异并不重要，因为在底层，data.frames 是建立在列表上的。尝试用这个替换vals：vals <- lapply(ll, function(x) x[[2]])，这是在答案中编写调用的另一种方式