应用vs地图：使用r中的tidyverse用映射使用tidyverse can filter list-list-list-lastumn数据框

Question

我正在尝试过滤data data_split _split _列中的行，基于data_split _split _和my_perts_df _ 。

因此，具体来说，我想：

1. 通过排除任何pert_iname in exclude在中过滤my_perts_df _表
2. ，然后根据grouping_var exclude /code>，可以是pert_class或pert_iname
3. 最后，使用此过滤的表来过滤data data_split_ 列/代码>表。

这是两个tibbles：

my_perts_df_ <- structure(list(pert_iname = c("DMSO", "gsk126", "jq1-s", "unc-0646", 
"geldanamycin", "decitabine", "gsk-j4", "vorinostat", "ly-294002", 
"staurosporine", "ly-294002", "kn-62", "ruxolitinib", "losmapimod", 
"ar a014418", "tofacitinib", "sp600125"), pert_class = c("control", 
"Epigenetic", "Epigenetic", "Epigenetic", "Epigenetic", "Epigenetic", 
"Epigenetic", "Epigenetic", "Epigenetic", "Kinase inhibitor", 
"Kinase inhibitor", "Kinase inhibitor", "Kinase inhibitor", "Kinase inhibitor", 
"Kinase inhibitor", "Kinase inhibitor", "Kinase inhibitor"), 
    P100 = c("P100", "P100", "P100", "P100", "P100", "P100", 
    "P100", "P100", "P100", "P100", "P100", "P100", "P100", "P100", 
    "P100", "P100", "P100"), GCP = c("GCP", "GCP", "GCP", "GCP", 
    "GCP", "GCP", "GCP", "GCP", "GCP", "GCP", "GCP", "GCP", NA, 
    "GCP", "GCP", "GCP", "GCP"), keep = c(TRUE, TRUE, TRUE, TRUE, 
    TRUE, TRUE, TRUE, TRUE, TRUE, TRUE, TRUE, TRUE, TRUE, TRUE, 
    TRUE, TRUE, TRUE)), row.names = c(NA, -17L), class = c("tbl_df", 
"tbl", "data.frame"))

data_split_ <- structure(list(dataset_type = c("P100", "P100", "GCP", "GCP"), 
    grouping_var = c("pert_class", "pert_iname", "pert_class", 
    "pert_iname"), filter_vars = list("Kinase inhibitor", "DMSO", 
        "Epigenetic", "DMSO"), exclude = list(NA_character_, 
        NA_character_, NA_character_, NA_character_), output_dir = c("output_final/p100", 
    "output_final/p100", "output_final/gcp", "output_final/gcp"
    ), data = list(structure(list(replicate_id = c("NPC--cabozantinib--VEGFR inhibitor::E01_acq_01::P-0063", 
    "PC3--DMSO--control::A02_acq_01::P-0034", "PC3--bms-345541--Kinase inhibitor::E12_acq_01::P-0034", 
    "MCF7--unc1215--Epigenetic::G09_DIA_acq_01::P-0015", "PC3--dasatinib--CML TKI::B05_acq_01::P-0061", 
    NA, "HUVEC--okadaic acid--Other::F11_acq_01::P-0057", "HUVEC--decitabine--Epigenetic::A05_acq_01::P-0038", 
    "PC3--ruxolitinib--Kinase inhibitor::E02_acq_01::P-0024", 
    "NPC--pd-0332991--Kinase inhibitor::B05_acq_02::P-0027", 
    "A549--vorinostat--Epigenetic::H05_acq_02::P-0058", "YAPC--dasatinib--CML TKI::B06_acq_01::P-0062", 
    "NPC--vorinostat--Epigenetic::H05_DIA_acq_01::P-0020", "A375--tbb--Kinase inhibitor::B10_acq_01::P-0022", 
    "NPC--nilotinib--CML TKI::B11_acq_03::P-0027", "A549--vandetanib--VEGFR inhibitor::E10_acq_01::P-0058", 
    "HUVEC--eplerenone--CV::D02_acq_01::P-0038", NA, "PC3--afuresertib--Kinase inhibitor::G08_acq_01::P-0034", 
    "NPC--ponatinib--CML TKI::A12_acq_01::P-0063", "PC3--sotrastaurin--IMiD::F10_acq_01::P-0024", 
    "HUVEC--atorvastatin--CV::A05_acq_01::P-0057", "HUVEC--losmapimod--Kinase inhibitor::C11_acq_01::P-0038", 
    "HUVEC--aspirin--CV::CS20180418_P100_HUVEC3_Batch2_P-0069_B11_acq_01::P-0069", 
    "A375--belinostat--Epigenetic::D08_DIA_acq_01::P-0017"), 
        pert_iname = c("cabozantinib", "DMSO", "bms-345541", 
        "unc1215", "dasatinib", "gefitinib", "okadaic acid", 
        "decitabine", "ruxolitinib", "pd-0332991", "vorinostat", 
        "dasatinib", "vorinostat", "tbb", "nilotinib", "vandetanib", 
        "eplerenone", "isoproterenol", "afuresertib", "ponatinib", 
        "sotrastaurin", "atorvastatin", "losmapimod", "aspirin", 
        "belinostat"), pert_class = c("VEGFR inhibitor", "control", 
        "Kinase inhibitor", "Epigenetic", "CML TKI", NA, "Other", 
        "Epigenetic", "Kinase inhibitor", "Kinase inhibitor", 
        "Epigenetic", "CML TKI", "Epigenetic", "Kinase inhibitor", 
        "CML TKI", "VEGFR inhibitor", "CV", NA, "Kinase inhibitor", 
        "CML TKI", "IMiD", "CV", "Kinase inhibitor", "CV", "Epigenetic"
        ), pr_gene_symbol = c("pS142 DPF2", "pS222 RBM17", "pS515 ZC3H14", 
        "pS235 RPS6", "pS405 C13orf8", "pS403 RNF169", "pS1035 GPATCH8", 
        "pS1075 AP1GBP1", "pS12 EIF4A3", "pS56 NOC2L", "pS230 SH3KBP1", 
        "pS2218 MAP4", "pS275 NANS", "pS405 C13orf8", "pS75 DDX54", 
        "pS446 BRAF", "pS500 C17orf85", "pS692 FAM129B", "pT596 MARK2", 
        "pS163 MAP3K2", "pS386 PFKP", "pS75 DDX54", "pS652 NUFIP2", 
        "pS275 NANS", "pS207 FASN"), value = c(0.181722044944763, 
        0.6737335, 0.021798491, 0.777894914150238, 0.0691990852355957, 
        0.2050631, 0.19466971, -0.30280375, -0.0263868570327759, 
        -0.196122169494629, 0.0888378620147705, -0.07376492, 
        0.05050297, 0.309347808361053, -0.146979808807373, -0.0501706600189209, 
        0.20668465, -0.279913723468781, 0.15067881, -0.0937120914459229, 
        -0.129467606544495, -0.07893503, -0.09858775, 0.312014818191528, 
        -0.0723450779914856)), row.names = c(NA, -25L), class = c("tbl_df", 
    "tbl", "data.frame")), structure(list(replicate_id = c("MCF7--decitabine--Epigenetic::F02_DIA_acq_01::P-0015", 
    "PC3--kn-93--Kinase inhibitor::E07_acq_01::P-0024", "NPC--verapamil--CV::E06_acq_02::P-0063", 
    "A549--ar a014418--Kinase inhibitor::D12_acq_01::P-0033", 
    "A375--gsk525762a--Epigenetic::C01_DIA_acq_01::P-0017", "YAPC--gsk126--Epigenetic::A05_acq_01::P-0030", 
    "YAPC--trametinib--Kinase inhibitor::D10_acq_01::P-0062", 
    "NPC--pri-724--Other::E07_acq_03::P-0027", "PC3--gossypetin--Kinase inhibitor::F09_acq_01::P-0024", 
    "A549--jq1-s--Epigenetic::D03_DIA_acq_01::P-0019", NA, "HUVEC--calyculin a--Kinase inhibitor::D08_acq_01::P-0057", 
    "MCF7--imatinib--CML TKI::C08_acq_01::P-0060", "NPC--byl719--Kinase inhibitor::H07_acq_03::P-0027", 
    "MCF7--c646--Epigenetic::C07_acq_02::P-0023", "NPC--tbb--Kinase inhibitor::B10_DIA_acq_01::P-0020", 
    "YAPC--ms-275--Epigenetic::A07_acq_01::P-0030", "MCF7--ms-275--Epigenetic::A08_DIA_acq_01::P-0015", 
    "HUVEC--curcumin--Other::D08_acq_01::P-0038", "MCF7--etoposide--Other::B02_acq_01::P-0023", 
    "MCF7--unc-0321--Epigenetic::E06_DIA_acq_01::P-0015", "NPC--ex527--Epigenetic::C20150526_P-0016_NPC_T1_cmpds_P-0016_C11_DIA_acq_01::P-0016", 
    "A549--sch 900776--Kinase inhibitor::B08_acq_01::P-0033", 
    "A549--gsk126--Epigenetic::A06_DIA_acq_01::P-0019", NA), 
        pert_iname = c("decitabine", "kn-93", "verapamil", "ar a014418", 
        "gsk525762a", "gsk126", "trametinib", "pri-724", "gossypetin", 
        "jq1-s", "cyclosporine", "calyculin a", "imatinib", "byl719", 
        "c646", "tbb", "ms-275", "ms-275", "curcumin", "etoposide", 
        "unc-0321", "ex527", "sch 900776", "gsk126", "ceritinib"
        ), pert_class = c("Epigenetic", "Kinase inhibitor", "CV", 
        "Kinase inhibitor", "Epigenetic", "Epigenetic", "Kinase inhibitor", 
        "Other", "Kinase inhibitor", "Epigenetic", NA, "Kinase inhibitor", 
        "CML TKI", "Kinase inhibitor", "Epigenetic", "Kinase inhibitor", 
        "Epigenetic", "Epigenetic", "Other", "Other", "Epigenetic", 
        "Epigenetic", "Kinase inhibitor", "Epigenetic", NA), 
        pr_gene_symbol = c("pS207 FASN", "pS230 SH3KBP1", "pS465 WDR20", 
        "pS652 NUFIP2", "pS601 LARP5", "pT3893 PLEC1_1", "pY321 DYRK1A", 
        "pS163 MAP3K2", "pS2 PAK2", "pS200 FOSL2_1", "pS402 SRRM1", 
        "pS2 TMSB4X", "pS1012 NUP214", "pS1075 AP1GBP1", "pS1219 BAT2", 
        "pS353 SRRM2", "pS142 DPF2", "pT2675 BAT2D1", "pS1035 GPATCH8", 
        "pS275 NANS", "pS556 ULK1", "pS103 DHX16", "pS515 ZC3H14", 
        "pS1075 AP1GBP1", "pS207 FASN"), value = c(0.0880586504936218, 
        -0.1592857837677, -0.232870101928711, 0.145888328552246, 
        -0.233265995979309, 0.659018278121948, 0.03665614, -0.080927848815918, 
        0.115837097167969, 0.234739303588867, 0.192527055740356, 
        0, 0.315743684768677, -0.426863610744476, -0.0367576479911804, 
        -0.16833115, 0.579586148262024, 0.263529172516428, -0.7783067, 
        -0.214210987091064, 0.105207920074463, 0.728229999542236, 
        -0.03392493724823, -0.106773972511292, -0.0157837867736816
        )), row.names = c(NA, -25L), class = c("tbl_df", "tbl", 
    "data.frame")), structure(list(replicate_id = c(NA, "A375--tretinoin--Other::A06_acq_01::G-0022", 
    "HAoSMC--dexamethasone--Other::A08_acq_01::G-0039a", "YAPC--epz-5687--Epigenetic::T20151105_Plate30_YAPC_T1_G-0030_G09_acq_01::G-0030", 
    "NPC--tbb--Kinase inhibitor::B12_acq_01::G-0020", "PC3--vandetanib--VEGFR inhibitor::E12_acq_02::G-0061", 
    "A549--DMSO--control::A01_acq_01::G-0033", "YAPC--flavopiridol--Kinase inhibitor::A04_acq_01::G-0032", 
    "HUVEC--tretinoin--Other::B09_acq_01::G-0038", "A375--sotrastaurin--IMiD::T_F12_acq_02::G-0022", 
    "A375--afuresertib--Kinase inhibitor::G07_acq_01::G-0028", 
    "MCF7--vx-970--Kinase inhibitor::C09_acq_02::G-0029", "A375--axitinib--VEGFR inhibitor::C01_acq_01::G-0059", 
    "NPC--trichostatin a--Epigenetic::C07_acq_02::G-0016R", "MCF7--nilotinib--CML TKI::B10_acq_01::G-0029", 
    "A549--DMSO--control::A01_acq_01::G-0058", "NPC--vx-970--Kinase inhibitor::C07_acq_01::G-0027", 
    "A375--tacrolimus--Other::H01_acq_01::G-0022", "A375--epz004777--Epigenetic::H01_acq_01::G-0017", 
    "A375--axitinib--VEGFR inhibitor::C01_acq_01::G-0059", "NPC--axitinib--VEGFR inhibitor::C01_acq_01::G-0063", 
    "HAoSMC--gsk126--Epigenetic::B05_acq_01::G-0039a", "NPC--ly-294002--Epigenetic::H12_acq_01::G-0016R", 
    "PC3--vemurafenib--Kinase inhibitor::E03_acq_01::G-0034", 
    "A375--lenalidomide--IMiD::C12_acq_01::G-0028"), pert_iname = c("olmesartan", 
    "tretinoin", "dexamethasone", "epz-5687", "tbb", "vandetanib", 
    "DMSO", "flavopiridol", "tretinoin", "sotrastaurin", "afuresertib", 
    "vx-970", "axitinib", "trichostatin a", "nilotinib", "DMSO", 
    "vx-970", "tacrolimus", "epz004777", "axitinib", "axitinib", 
    "gsk126", "ly-294002", "vemurafenib", "lenalidomide"), pert_class = c(NA, 
    "Other", "Other", "Epigenetic", "Kinase inhibitor", "VEGFR inhibitor", 
    "control", "Kinase inhibitor", "Other", "IMiD", "Kinase inhibitor", 
    "Kinase inhibitor", "VEGFR inhibitor", "Epigenetic", "CML TKI", 
    "control", "Kinase inhibitor", "Other", "Epigenetic", "VEGFR inhibitor", 
    "VEGFR inhibitor", "Epigenetic", "Epigenetic", "Kinase inhibitor", 
    "IMiD"), pr_gene_symbol = c("H3K9me2S10ph1K14ac0", "H3K27me1K36me3", 
    "H3K27me2K36me2", "H3K9me2S10ph1K14ac0", "H3K27me3K36me1", 
    "H4(4to17)K5ac1K8ac1K12ac1K16ac1me0", "H3K27me3K36me2", "H3K27me2K36me1", 
    "H3K9ac1S10ph1K14ac1", "H3K9me1K14ac1", "H3K27me2K36me2", 
    "H3K9me1S10ph1K14ac0", "H3K9me1K14ac0", "H3K9me1S10ph1K14ac1", 
    "H3K27me2K36me2", "H3K27me2K36me1", "H3K27me3K36me0", "H3K27ac1K36me0", 
    "H3K9me3S10ph1K14ac0", "H3K9me2S10ph1K14ac1", "H4(20to23)K20me3", 
    "H3NORM(41-49)", "H3K18ac0K23ac1", "H3K27ac1K36me0", "H3K27me3K36me2"
    ), value = c(0.0708658695220947, -0.27732902765274, 0.335925102233887, 
    0.148061037063599, 0.0192266702651978, -0.06201318, -0.32350385, 
    -0.113838315010071, 0.888034820556641, 0.0549858212471008, 
    -0.474118649959564, -0.0354797840118408, -0.1448524, -1.6616924, 
    -0.0739709138870239, -0.191317439079285, 0.404392004013062, 
    -1.12328338623047, -0.106091976, -0.45647645, 0.166058421134949, 
    -0.0442338585853577, -0.59617877, -0.9624169, 0.0658320188522339
    )), row.names = c(NA, -25L), class = c("tbl_df", "tbl", "data.frame"
    )), structure(list(replicate_id = c("NPC--pd0325901--Kinase inhibitor::C02_acq_01::G-0027", 
    "NPC--resveratrol--Other::B10_acq_02::G-0016R", "HAoSMC--sirolimus--Other::F04_acq_01::G-0039a", 
    "PC3--rgfp966--Epigenetic::E12_acq_01::G-0024", "MCF7--rolipram--Other::D10_acq_01::G-0023", 
    "PC3--sirolimus--Other::F11_acq_01::G-0018", "A375--sorafenib--VEGFR inhibitor::C10_acq_01::G-0059", 
    "PC3--tbb--Kinase inhibitor::B10_acq_01::G-0024", NA, "MCF7--staurosporine--Kinase inhibitor::G11_acq_01::G-0029", 
    "A375--pravastatin--CV::H02_acq_01::G-0028", "PC3--selumetinib--Kinase inhibitor::B01_acq_01::G-0034", 
    NA, "A549--sp600125--Kinase inhibitor::G04_acq_01::G-0033", 
    "A549--salermide--Epigenetic::D11_acq_01::G-0019", "MCF7--roscovitine--Kinase inhibitor::C11_acq_01::G-0023", 
    "YAPC--everolimus--Kinase inhibitor::D02_acq_01::G-0032", 
    "YAPC--verteporfin--Other::F06_acq_02::G-0032", "PC3--osi-027--Kinase inhibitor::A12_acq_02::G-0018", 
    "A549--dexamethasone--Other::D04_acq_01::G-0025", "HAoSMC--jq1-s--Epigenetic::G03_acq_01::G-0039a", 
    "HUVEC--tofacitinib--Kinase inhibitor::H08_acq_01::G-0038", 
    "MCF7--methylstat--Epigenetic::E09_acq_02::G-0015", NA, "PC3--DMSO--control::A03_acq_01::G-0018"
    ), pert_iname = c("pd0325901", "resveratrol", "sirolimus", 
    "rgfp966", "rolipram", "sirolimus", "sorafenib", "tbb", "lapatinib", 
    "staurosporine", "pravastatin", "selumetinib", "ceritinib", 
    "sp600125", "salermide", "roscovitine", "everolimus", "verteporfin", 
    "osi-027", "dexamethasone", "jq1-s", "tofacitinib", "methylstat", 
    "lapatinib", "DMSO"), pert_class = c("Kinase inhibitor", 
    "Other", "Other", "Epigenetic", "Other", "Other", "VEGFR inhibitor", 
    "Kinase inhibitor", NA, "Kinase inhibitor", "CV", "Kinase inhibitor", 
    NA, "Kinase inhibitor", "Epigenetic", "Kinase inhibitor", 
    "Kinase inhibitor", "Other", "Kinase inhibitor", "Other", 
    "Epigenetic", "Kinase inhibitor", "Epigenetic", NA, "control"
    ), pr_gene_symbol = c("H3K9me0S10ph1K14ac0", "H3K27me2K36me1", 
    "H3K9me2K14ac0", "H3NORM(41-49)", "H3K4me2", "H3K27me0K36me0", 
    "H3K9me1K14ac1", "H3K9ac1K14ac1", "H4(4to17)K5ac1me0", "H3K18ac0K23ac0", 
    "H3K4me2", "H3K4me1", "H3K27me3K36me0", "H3K27me2K36me2", 
    "H3K9me2K14ac0", "H3K27me1K36me0", "H3K27me3K36me2", "H3K27ac1K36me0", 
    "H3K18ac1K23ac1", "H3K79me0", "H3K27me3K36me1", "H3K27me3K36me2", 
    "H3K27me3K36me0", "H4(4to17)K8ac1K12ac1me0", "H3K4me1"), 
        value = c(-0.601195812225342, -0.056107312, 0.131532583385706, 
        -0.121334552764893, -0.216784358024597, -0.261847794055939, 
        -0.218853, -0.189859390258789, 0.0388565063476562, -0.322605133056641, 
        -0.0750547647476196, 0, -0.29474682, 0.3334179, -0.381944835186005, 
        -0.677833437919617, 0.150129675865173, 0.153744220733643, 
        1.26011848449707, 0.225029110908508, 0.124394655227661, 
        0.0346214175224304, 0.271708130836487, 0.57789421081543, 
        -0.622473895549774)), row.names = c(NA, -25L), class = c("tbl_df", 
    "tbl", "data.frame")))), row.names = c(NA, -4L), class = c("tbl_df", 
"tbl", "data.frame"))

到目前为止，这是我的代码：

 data_split_ %>%
  mutate(perturbation_filtered_data = pmap(.l = list(grouping_var, filter_vars, exclude, data), .f = function(g_var, f_var, e_var, d){
    
    if (any(is.na(unlist(e_var)))) {
      exclude_term <- ""
    } else {
      exclude_term <- unlist(e_var)
    }
    
    new_pert_df <- my_perts_df_ %>%
      filter(!(pert_iname %in% exclude_term)) %>%
      filter(sym(g_var) == f_var)
    res <- d %>% 
      filter(pert_iname %in% new_pert_df$pert_iname)
    return(res)
  }))

# A tibble: 4 x 7
  dataset_type grouping_var filter_vars exclude   output_dir        data              perturbation_filtered_data
  <chr>        <chr>        <list>      <list>    <chr>             <list>            <list>                    
1 P100         pert_class   <chr [1]>   <chr [1]> output_final/p100 <tibble [25 x 5]> <tibble [0 x 5]>          
2 P100         pert_iname   <chr [1]>   <chr [1]> output_final/p100 <tibble [25 x 5]> <tibble [0 x 5]>          
3 GCP          pert_class   <chr [1]>   <chr [1]> output_final/gcp  <tibble [25 x 5]> <tibble [0 x 5]>          
4 GCP          pert_iname   <chr [1]>   <chr [1]> output_final/gcp  <tibble [25 x 5]> <tibble [0 x 5]>          

过滤的数据集都带有0行回来，这是错误的。我期望每个数据集中有几个保留行。任何帮助都非常感谢！