Skip to contents

Analysis

Source: vignettes/analysis.qmd

Preliminary 

## load R package
library(msDiaLogue)
## preprocessing
fileName <- "../inst/extdata/Toy_Spectronaut_Data.csv"
dataSet <- preprocessing(fileName,
                         filterNaN = TRUE, filterUnique = 2,
                         replaceBlank = TRUE, saveRm = TRUE)
## transformation
dataTran <- transform(dataSet, logFold = 2)
## annotation-based filtering
dataFiltAnno <- filterOutIn(dataTran, listName = "ALBU_BOVIN",
                            removeList = TRUE, saveRm = TRUE)
## normalization
dataNorm <- normalize(dataFiltAnno, normalizeType = "median")
## data-driven filtering
dataFilt <- filterNA(dataNorm, minProp = 0.51, by = "cond", saveRm = TRUE)
## imputation
dataImput <- impute.min_local(dataFilt)
## summarization
dataSumm <- summarize(dataImput, saveSumm = TRUE)

The functions in the analysis module calculate the results that can be used in subsequent visualizations.

Note: The following analyses compare all other conditions against the reference condition, which is specified by the argument ref, for multiple comparisons. If ref is not provided, it will be automatically generated by all the combinations of two conditions, based on the level attributes of the condition.
For example, suppose there are three conditions in the data: “A”, “B”, and “C”. If you specify ref = "A", then the result includes two comparisons: “B-A” and “C-A”. If ref = NULL, there will be three comparisons: “A-B”, “A-C”, and “B-C”.

Student’s t-test

Example 

anlys_t <- analyze.t(dataImput, ref = "50fmol", adjust.method = "none",
                     saveRes = TRUE)
Note: In the Student’s t-test, a warning message might appear, stating “Data are essentially constant,” which means that the data contain proteins with the same value in all samples. In this case, the p-value of t-test returns NaN. 
#> $`100fmol-50fmol`
NUD4B_HUMAN (+1) A0A7P0T808_HUMAN (+1) A0A8I5KU53_HUMAN (+1) ZN840_HUMAN CC85C_HUMAN C9JEV0_HUMAN (+1) C9JNU9_HUMAN CYC_BOVIN TRFE_BOVIN F8W0H2_HUMAN H0Y7V7_HUMAN (+1) H0YD14_HUMAN H3BUF6_HUMAN H7C1W4_HUMAN (+1) H7C3M7_HUMAN TLR3_HUMAN LRIG2_HUMAN RAB3D_HUMAN ADH1_YEAST LYSC_CHICK BGAL_ECOLI CYTA_HUMAN KPCB_HUMAN LIPL_HUMAN CO6_HUMAN BGAL_HUMAN SYTC_HUMAN CASPE_HUMAN DCAF6_HUMAN DALD3_HUMAN HGNAT_HUMAN RFFL_HUMAN RN185_HUMAN ZN462_HUMAN ALKB7_HUMAN POLK_HUMAN ACAD8_HUMAN
100fmol mean 0.8970585 2.1918625 1.1190602 -1.3091319 -0.7616711 -2.5823326 -1.3720362 3.5549310 4.1598999 -0.4095082 0.5375144 0.4206618 -1.5899192 -1.7349727 0.3759821 -1.9399514 -0.9633006 0.5924669 5.0671570 3.9307808 4.8065148 -0.6487769 0.4907685 -1.2108609 -2.6386252 2.1157812 4.8885237 -3.0010370 -1.6555058 0.1042002 0.8543029 -2.6865201 0.4758179 -0.8105107 -2.4501697 -1.7550160 -1.8485990
50fmol mean 1.5018335 -1.4867645 0.7414668 -1.2046036 -0.6978038 -0.7687851 -0.7991065 3.7761356 4.5984741 0.1523311 0.2432124 1.2303125 -0.7489724 -2.6300921 0.6353621 -1.8284465 -0.9168044 1.2127423 5.0713696 3.7915558 4.9848046 1.2366567 1.1905031 -0.7024215 -1.8418109 3.0118366 5.6824480 -0.7531886 -1.7462293 -0.4885937 0.7407263 -2.7704565 0.0157819 -1.0501134 -1.3023443 -2.7093144 -2.2373695
difference -0.6047750 3.6786271 0.3775934 -0.1045283 -0.0638672 -1.8135474 -0.5729297 -0.2212046 -0.4385743 -0.5618393 0.2943020 -0.8096507 -0.8409468 0.8951194 -0.2593801 -0.1115049 -0.0464961 -0.6202754 -0.0042126 0.1392250 -0.1782898 -1.8854336 -0.6997346 -0.5084394 -0.7968143 -0.8960554 -0.7939243 -2.2478484 0.0907235 0.5927939 0.1135766 0.0839364 0.4600360 0.2396026 -1.1478254 0.9542984 0.3887704
p-value 0.0581502 0.0388908 0.6931358 0.8308249 0.7868695 0.0166976 0.1697253 0.3424744 0.0542313 0.0467543 0.2253282 0.0113872 0.0012974 0.2251108 0.3504772 0.6557080 0.8956558 0.0027392 0.9732646 0.4130959 0.3971972 0.0119719 0.0393259 0.1523881 0.0125963 0.0009685 0.0078059 0.0042206 0.7694137 0.1562143 0.7964029 0.8762626 0.1583942 0.8080628 0.0289793 0.3258409 0.1442833 
#> $`200fmol-50fmol`
NUD4B_HUMAN (+1) A0A7P0T808_HUMAN (+1) A0A8I5KU53_HUMAN (+1) ZN840_HUMAN CC85C_HUMAN C9JEV0_HUMAN (+1) C9JNU9_HUMAN CYC_BOVIN TRFE_BOVIN F8W0H2_HUMAN H0Y7V7_HUMAN (+1) H0YD14_HUMAN H3BUF6_HUMAN H7C1W4_HUMAN (+1) H7C3M7_HUMAN TLR3_HUMAN LRIG2_HUMAN RAB3D_HUMAN ADH1_YEAST LYSC_CHICK BGAL_ECOLI CYTA_HUMAN KPCB_HUMAN LIPL_HUMAN CO6_HUMAN BGAL_HUMAN SYTC_HUMAN CASPE_HUMAN DCAF6_HUMAN DALD3_HUMAN HGNAT_HUMAN RFFL_HUMAN RN185_HUMAN ZN462_HUMAN ALKB7_HUMAN POLK_HUMAN ACAD8_HUMAN
200fmol mean 1.2129798 0.9476617 1.6646748 -0.8707864 -0.9704449 -2.4355408 -1.0109247 4.8036461 4.9771724 -0.1293629 0.8123346 0.9012464 -1.0788546 -1.6920949 0.8326191 -1.8320831 -0.7943162 0.7906959 6.301899 5.1797871 6.103709 -0.3601495 0.8966850 -0.7324823 -2.4259129 2.4717449 5.4536305 -3.2541540 -1.7246925 0.7037732 1.1711891 -2.3300528 -0.0848047 -0.4918852 -1.5365371 -2.0645198 -1.4177061
50fmol mean 1.5018335 -1.4867645 0.7414668 -1.2046036 -0.6978038 -0.7687851 -0.7991065 3.7761356 4.5984741 0.1523311 0.2432124 1.2303125 -0.7489724 -2.6300921 0.6353621 -1.8284465 -0.9168044 1.2127423 5.071370 3.7915558 4.984805 1.2366567 1.1905031 -0.7024215 -1.8418109 3.0118366 5.6824480 -0.7531886 -1.7462293 -0.4885937 0.7407263 -2.7704565 0.0157819 -1.0501134 -1.3023443 -2.7093144 -2.2373695
difference -0.2888537 2.4344262 0.9232081 0.3338172 -0.2726410 -1.6667557 -0.2118181 1.0275105 0.3786983 -0.2816940 0.5691222 -0.3290661 -0.3298822 0.9379972 0.1972570 -0.0036365 0.1224882 -0.4220464 1.230530 1.3882313 1.118905 -1.5968062 -0.2938181 -0.0300608 -0.5841020 -0.5400917 -0.2288175 -2.5009654 0.0215368 1.1923669 0.4304628 0.4404038 -0.1005866 0.5582282 -0.2341928 0.6447946 0.8196634
p-value 0.2025274 0.1434748 0.3464224 0.4440254 0.5655557 0.0255350 0.5895934 0.0007935 0.0883691 0.0841731 0.0291135 0.1012709 0.0855658 0.2194615 0.4621356 0.9816929 0.6312976 0.0230185 0.002502 0.0001215 0.000688 0.0198007 0.1608123 0.9033510 0.0126935 0.0234382 0.2567351 0.0047086 0.9473425 0.0034473 0.4622600 0.4081276 0.7757221 0.5683170 0.5904546 0.4096305 0.2225415

Details

The Student’s t-test is used to compare the means between two conditions for each protein, reporting both the difference in means between the conditions and the p-value of the test.

The argument adjust.method is used to specify the testing correction procedure to be applied to p-values. This adjustment is very common in DNA or RNA-Seq analyses, where datasets are very large and where researchers are most interested in controlling the Type I error rate when conducting multiple comparisons.

However, for mass spectrometry-based proteomics results, the dataset sizes are smaller than in sequencing analyses, and testing corrections can be too harsh of a threshold to apply. Most often, applying any testing correction to proteomics data results in there being zero significant changes. This does not mean that nothing is meaningfully changing in your dataset. It does mean that these corrections are usually not a useful tool for finding biologically-relevant changes in your dataset.

Also keep in mind that reducing Type I error typically comes at the cost of increasing Type II error, and vice versa. There is no way to eliminate all error; each researcher must decide whether they are more comfortable with having more false positives or more false negatives in the dataset, and choose their analysis strategies accordingly.

UConn PMF recommends not applying testing corrections to your proteomics dataset, but if you would like to explore the effects of doing so, several methods are provided below:

  1. “BH” or its alias “fdr”: Benjamini and Hochberg (1995).

  2. “BY”: Benjamini and Yekutieli (2001).

  3. “bonferroni”: Bonferroni (1936).

  4. “hochberg”: Hochberg (1988).

  5. “holm”: Holm (1979).

  6. “hommel”: Hommel (1988).

Each method offers its own balance between statistical power and error control. The default value "none" indicates that no correction is applied.

Empirical Bayes moderated t-test

Example 

anlys_modt <- analyze.mod_t(dataImput, ref = "50fmol", adjust.method = "none",
                            saveRes = TRUE)
Note: In the moderated t-test, a warning message might occur stating, “Zero sample variances detected, have been offset away from zero.” This warning corresponds to examples of proteins that exhibited identical quant values, either pre- or post-imputation, and therefore no variance is present across conditions for those proteins. This does not impede downstream analysis; it merely serves to alert users to its occurrence. 
#> $`100fmol-50fmol`
NUD4B_HUMAN (+1) A0A7P0T808_HUMAN (+1) A0A8I5KU53_HUMAN (+1) ZN840_HUMAN CC85C_HUMAN C9JEV0_HUMAN (+1) C9JNU9_HUMAN CYC_BOVIN TRFE_BOVIN F8W0H2_HUMAN H0Y7V7_HUMAN (+1) H0YD14_HUMAN H3BUF6_HUMAN H7C1W4_HUMAN (+1) H7C3M7_HUMAN TLR3_HUMAN LRIG2_HUMAN RAB3D_HUMAN ADH1_YEAST LYSC_CHICK BGAL_ECOLI CYTA_HUMAN KPCB_HUMAN LIPL_HUMAN CO6_HUMAN BGAL_HUMAN SYTC_HUMAN CASPE_HUMAN DCAF6_HUMAN DALD3_HUMAN HGNAT_HUMAN RFFL_HUMAN RN185_HUMAN ZN462_HUMAN ALKB7_HUMAN POLK_HUMAN ACAD8_HUMAN
100fmol mean 0.8970585 2.1918625 1.1190602 -1.3091319 -0.7616711 -2.5823326 -1.3720362 3.5549310 4.1598999 -0.4095082 0.5375144 0.4206618 -1.5899192 -1.7349727 0.3759821 -1.9399514 -0.9633006 0.5924669 5.0671570 3.9307808 4.8065148 -0.6487769 0.4907685 -1.2108609 -2.6386252 2.1157812 4.8885237 -3.0010370 -1.6555058 0.1042002 0.8543029 -2.6865201 0.4758179 -0.8105107 -2.4501697 -1.7550160 -1.8485990
50fmol mean 1.5018335 -1.4867645 0.7414668 -1.2046036 -0.6978038 -0.7687851 -0.7991065 3.7761356 4.5984741 0.1523311 0.2432124 1.2303125 -0.7489724 -2.6300921 0.6353621 -1.8284465 -0.9168044 1.2127423 5.0713696 3.7915558 4.9848046 1.2366567 1.1905031 -0.7024215 -1.8418109 3.0118366 5.6824480 -0.7531886 -1.7462293 -0.4885937 0.7407263 -2.7704565 0.0157819 -1.0501134 -1.3023443 -2.7093144 -2.2373695
difference -0.6047750 3.6786271 0.3775934 -0.1045283 -0.0638672 -1.8135474 -0.5729297 -0.2212046 -0.4385743 -0.5618393 0.2943020 -0.8096507 -0.8409468 0.8951194 -0.2593801 -0.1115049 -0.0464961 -0.6202754 -0.0042126 0.1392250 -0.1782898 -1.8854336 -0.6997346 -0.5084394 -0.7968143 -0.8960554 -0.7939243 -2.2478484 0.0907235 0.5927939 0.1135766 0.0839364 0.4600360 0.2396026 -1.1478254 0.9542984 0.3887704
p-value 0.0214224 0.0038262 0.5703100 0.7796763 0.8594470 0.0008250 0.1203648 0.3250638 0.0522133 0.0188337 0.2188577 0.0028155 0.0007490 0.0958207 0.3082681 0.6385995 0.8794846 0.0044135 0.9837690 0.4809873 0.3791024 0.0001446 0.0122187 0.0764326 0.0024990 0.0005587 0.0026648 0.0000483 0.7698904 0.0584241 0.7965475 0.8342575 0.1497018 0.7258645 0.0045167 0.1648238 0.3822198 
#> $`200fmol-50fmol`
NUD4B_HUMAN (+1) A0A7P0T808_HUMAN (+1) A0A8I5KU53_HUMAN (+1) ZN840_HUMAN CC85C_HUMAN C9JEV0_HUMAN (+1) C9JNU9_HUMAN CYC_BOVIN TRFE_BOVIN F8W0H2_HUMAN H0Y7V7_HUMAN (+1) H0YD14_HUMAN H3BUF6_HUMAN H7C1W4_HUMAN (+1) H7C3M7_HUMAN TLR3_HUMAN LRIG2_HUMAN RAB3D_HUMAN ADH1_YEAST LYSC_CHICK BGAL_ECOLI CYTA_HUMAN KPCB_HUMAN LIPL_HUMAN CO6_HUMAN BGAL_HUMAN SYTC_HUMAN CASPE_HUMAN DCAF6_HUMAN DALD3_HUMAN HGNAT_HUMAN RFFL_HUMAN RN185_HUMAN ZN462_HUMAN ALKB7_HUMAN POLK_HUMAN ACAD8_HUMAN
200fmol mean 1.2129798 0.9476617 1.6646748 -0.8707864 -0.9704449 -2.4355408 -1.0109247 4.8036461 4.9771724 -0.1293629 0.8123346 0.9012464 -1.0788546 -1.6920949 0.8326191 -1.8320831 -0.7943162 0.7906959 6.3018992 5.1797871 6.103709 -0.3601495 0.8966850 -0.7324823 -2.4259129 2.4717449 5.4536305 -3.2541540 -1.7246925 0.7037732 1.1711891 -2.3300528 -0.0848047 -0.4918852 -1.5365371 -2.0645198 -1.4177061
50fmol mean 1.5018335 -1.4867645 0.7414668 -1.2046036 -0.6978038 -0.7687851 -0.7991065 3.7761356 4.5984741 0.1523311 0.2432124 1.2303125 -0.7489724 -2.6300921 0.6353621 -1.8284465 -0.9168044 1.2127423 5.0713696 3.7915558 4.984805 1.2366567 1.1905031 -0.7024215 -1.8418109 3.0118366 5.6824480 -0.7531886 -1.7462293 -0.4885937 0.7407263 -2.7704565 0.0157819 -1.0501134 -1.3023443 -2.7093144 -2.2373695
difference -0.2888537 2.4344262 0.9232081 0.3338172 -0.2726410 -1.6667557 -0.2118181 1.0275105 0.3786983 -0.2816940 0.5691222 -0.3290661 -0.3298822 0.9379972 0.1972570 -0.0036365 0.1224882 -0.4220464 1.2305295 1.3882313 1.118905 -1.5968062 -0.2938181 -0.0300608 -0.5841020 -0.5400917 -0.2288175 -2.5009654 0.0215368 1.1923669 0.4304628 0.4404038 -0.1005866 0.5582282 -0.2341928 0.6447946 0.8196634
p-value 0.2312199 0.0359386 0.1789407 0.3787725 0.4547851 0.0015617 0.5482649 0.0004443 0.0875741 0.1992163 0.0273178 0.1551682 0.1050869 0.0825627 0.4342264 0.9877217 0.6904245 0.0350320 0.0000531 0.0000094 0.000090 0.0005884 0.2403498 0.9107316 0.0164315 0.0161169 0.2996464 0.0000173 0.9445437 0.0011949 0.3373357 0.2836480 0.7424239 0.4192566 0.4918936 0.3374121 0.0797911

Details

The main distinction between the Student’s and empirical Bayes moderated t-tests (Smyth 2004) lies in how variance is computed. While the Student’s t-test calculates variance based on the data available for each protein individually (which will be limited by the number of replicates included for each condition), the moderated t-test utilizes information from all replicates of every protein in the current dataset to calculate variance.

Wilcoxon test

Example 

anlys_wilcox <- analyze.wilcox(dataImput, ref = "50fmol", adjust.method = "none",
                               saveRes = TRUE)
Note: In the Wilcoxon test, the warning message “cannot compute exact p-value with ties.” may be displayed. This warning means that some values with tied rankings and the sample size is lower than 50, which prevents the exact p-value from being calculated. In such cases, a normal approximation is used. If all samples for a protein have the same value, the corresponding p-value returns NaN. 
#> $`100fmol-50fmol`
NUD4B_HUMAN (+1) A0A7P0T808_HUMAN (+1) A0A8I5KU53_HUMAN (+1) ZN840_HUMAN CC85C_HUMAN C9JEV0_HUMAN (+1) C9JNU9_HUMAN CYC_BOVIN TRFE_BOVIN F8W0H2_HUMAN H0Y7V7_HUMAN (+1) H0YD14_HUMAN H3BUF6_HUMAN H7C1W4_HUMAN (+1) H7C3M7_HUMAN TLR3_HUMAN LRIG2_HUMAN RAB3D_HUMAN ADH1_YEAST LYSC_CHICK BGAL_ECOLI CYTA_HUMAN KPCB_HUMAN LIPL_HUMAN CO6_HUMAN BGAL_HUMAN SYTC_HUMAN CASPE_HUMAN DCAF6_HUMAN DALD3_HUMAN HGNAT_HUMAN RFFL_HUMAN RN185_HUMAN ZN462_HUMAN ALKB7_HUMAN POLK_HUMAN ACAD8_HUMAN
100fmol mean 0.8970585 2.1918625 1.1190602 -1.3091319 -0.7616711 -2.5823326 -1.3720362 3.5549310 4.1598999 -0.4095082 0.5375144 0.4206618 -1.5899192 -1.7349727 0.3759821 -1.9399514 -0.9633006 0.5924669 5.0671570 3.9307808 4.8065148 -0.6487769 0.4907685 -1.2108609 -2.6386252 2.1157812 4.8885237 -3.0010370 -1.6555058 0.1042002 0.8543029 -2.6865201 0.4758179 -0.8105107 -2.4501697 -1.7550160 -1.8485990
50fmol mean 1.5018335 -1.4867645 0.7414668 -1.2046036 -0.6978038 -0.7687851 -0.7991065 3.7761356 4.5984741 0.1523311 0.2432124 1.2303125 -0.7489724 -2.6300921 0.6353621 -1.8284465 -0.9168044 1.2127423 5.0713696 3.7915558 4.9848046 1.2366567 1.1905031 -0.7024215 -1.8418109 3.0118366 5.6824480 -0.7531886 -1.7462293 -0.4885937 0.7407263 -2.7704565 0.0157819 -1.0501134 -1.3023443 -2.7093144 -2.2373695
difference -0.6047750 3.6786271 0.3775934 -0.1045283 -0.0638672 -1.8135474 -0.5729297 -0.2212046 -0.4385743 -0.5618393 0.2943020 -0.8096507 -0.8409468 0.8951194 -0.2593801 -0.1115049 -0.0464961 -0.6202754 -0.0042126 0.1392250 -0.1782898 -1.8854336 -0.6997346 -0.5084394 -0.7968143 -0.8960554 -0.7939243 -2.2478484 0.0907235 0.5927939 0.1135766 0.0839364 0.4600360 0.2396026 -1.1478254 0.9542984 0.3887704
p-value 0.0571429 0.0285714 0.8857143 0.8857143 0.6857143 0.0285714 0.2000000 0.4857143 0.1142857 0.1142857 0.1714286 0.0285714 0.0285714 0.2000000 0.2000000 1.0000000 0.8285714 0.0285714 0.8857143 0.6857143 0.2000000 0.0285714 0.0285714 0.1142857 0.0285714 0.0285714 0.0285714 0.0285714 0.6857143 0.2857143 1.0000000 0.6857143 0.2000000 0.8857143 0.0285714 0.6857143 0.1142857 
#> $`200fmol-50fmol`
NUD4B_HUMAN (+1) A0A7P0T808_HUMAN (+1) A0A8I5KU53_HUMAN (+1) ZN840_HUMAN CC85C_HUMAN C9JEV0_HUMAN (+1) C9JNU9_HUMAN CYC_BOVIN TRFE_BOVIN F8W0H2_HUMAN H0Y7V7_HUMAN (+1) H0YD14_HUMAN H3BUF6_HUMAN H7C1W4_HUMAN (+1) H7C3M7_HUMAN TLR3_HUMAN LRIG2_HUMAN RAB3D_HUMAN ADH1_YEAST LYSC_CHICK BGAL_ECOLI CYTA_HUMAN KPCB_HUMAN LIPL_HUMAN CO6_HUMAN BGAL_HUMAN SYTC_HUMAN CASPE_HUMAN DCAF6_HUMAN DALD3_HUMAN HGNAT_HUMAN RFFL_HUMAN RN185_HUMAN ZN462_HUMAN ALKB7_HUMAN POLK_HUMAN ACAD8_HUMAN
200fmol mean 1.2129798 0.9476617 1.6646748 -0.8707864 -0.9704449 -2.4355408 -1.0109247 4.8036461 4.9771724 -0.1293629 0.8123346 0.9012464 -1.0788546 -1.6920949 0.8326191 -1.8320831 -0.7943162 0.7906959 6.3018992 5.1797871 6.1037095 -0.3601495 0.8966850 -0.7324823 -2.4259129 2.4717449 5.4536305 -3.2541540 -1.7246925 0.7037732 1.1711891 -2.3300528 -0.0848047 -0.4918852 -1.5365371 -2.0645198 -1.4177061
50fmol mean 1.5018335 -1.4867645 0.7414668 -1.2046036 -0.6978038 -0.7687851 -0.7991065 3.7761356 4.5984741 0.1523311 0.2432124 1.2303125 -0.7489724 -2.6300921 0.6353621 -1.8284465 -0.9168044 1.2127423 5.0713696 3.7915558 4.9848046 1.2366567 1.1905031 -0.7024215 -1.8418109 3.0118366 5.6824480 -0.7531886 -1.7462293 -0.4885937 0.7407263 -2.7704565 0.0157819 -1.0501134 -1.3023443 -2.7093144 -2.2373695
difference -0.2888537 2.4344262 0.9232081 0.3338172 -0.2726410 -1.6667557 -0.2118181 1.0275105 0.3786983 -0.2816940 0.5691222 -0.3290661 -0.3298822 0.9379972 0.1972570 -0.0036365 0.1224882 -0.4220464 1.2305295 1.3882313 1.1189049 -1.5968062 -0.2938181 -0.0300608 -0.5841020 -0.5400917 -0.2288175 -2.5009654 0.0215368 1.1923669 0.4304628 0.4404038 -0.1005866 0.5582282 -0.2341928 0.6447946 0.8196634
p-value 0.2000000 0.3142857 0.6857143 0.6285714 0.8857143 0.0285714 0.8285714 0.0285714 0.2000000 0.1142857 0.0571429 0.1714286 0.1142857 0.2000000 0.6857143 0.6285714 0.4285714 0.0571429 0.0285714 0.0285714 0.0285714 0.0285714 0.4571429 1.0000000 0.0285714 0.1142857 0.6857143 0.0285714 0.8857143 0.0285714 0.2000000 0.6857143 0.8857143 0.6857143 0.8857143 0.8857143 0.4857143

Details

The Wilcoxon test is a non-parametric alternative to the two-sample t-test. If paired = TRUE, a Wilcoxon signed-rank test is performed to test the null hypothesis that the distribution of the difference between the two conditions for the protein is symmetric about zero. If paired = FALSE, a Wilcoxon rank-sum test (also known as Mann-Whitney test) is performed to test the null hypothesis that the distribution of the two conditions for the protein are the same.

MA

Example 

anlys_ma <- analyze.ma(dataImput, ref = "50fmol", saveRes = TRUE)
#> $`100fmol-50fmol`
NUD4B_HUMAN (+1) A0A7P0T808_HUMAN (+1) A0A8I5KU53_HUMAN (+1) ZN840_HUMAN CC85C_HUMAN C9JEV0_HUMAN (+1) C9JNU9_HUMAN CYC_BOVIN TRFE_BOVIN F8W0H2_HUMAN H0Y7V7_HUMAN (+1) H0YD14_HUMAN H3BUF6_HUMAN H7C1W4_HUMAN (+1) H7C3M7_HUMAN TLR3_HUMAN LRIG2_HUMAN RAB3D_HUMAN ADH1_YEAST LYSC_CHICK BGAL_ECOLI CYTA_HUMAN KPCB_HUMAN LIPL_HUMAN CO6_HUMAN BGAL_HUMAN SYTC_HUMAN CASPE_HUMAN DCAF6_HUMAN DALD3_HUMAN HGNAT_HUMAN RFFL_HUMAN RN185_HUMAN ZN462_HUMAN ALKB7_HUMAN POLK_HUMAN ACAD8_HUMAN
100fmol mean 0.8970585 2.191863 1.1190602 -1.3091319 -0.7616711 -2.5823326 -1.3720362 3.5549310 4.1598999 -0.4095082 0.5375144 0.4206618 -1.5899192 -1.7349727 0.3759821 -1.9399514 -0.9633006 0.5924669 5.0671570 3.930781 4.8065148 -0.6487769 0.4907685 -1.2108609 -2.6386252 2.1157812 4.8885237 -3.0010370 -1.6555058 0.1042002 0.8543029 -2.6865201 0.4758179 -0.8105107 -2.450170 -1.7550160 -1.8485990
50fmol mean 1.5018335 -1.486765 0.7414668 -1.2046036 -0.6978038 -0.7687851 -0.7991065 3.7761356 4.5984741 0.1523311 0.2432124 1.2303125 -0.7489724 -2.6300921 0.6353621 -1.8284465 -0.9168044 1.2127423 5.0713696 3.791556 4.9848046 1.2366567 1.1905031 -0.7024215 -1.8418109 3.0118366 5.6824480 -0.7531886 -1.7462293 -0.4885937 0.7407263 -2.7704565 0.0157819 -1.0501134 -1.302344 -2.7093144 -2.2373695
A 1.1994460 0.352549 0.9302635 -1.2568677 -0.7297375 -1.6755588 -1.0855714 3.6655333 4.3791870 -0.1285885 0.3903634 0.8254872 -1.1694458 -2.1825324 0.5056721 -1.8841990 -0.9400525 0.9026046 5.0692633 3.861168 4.8956597 0.2939399 0.8406358 -0.9566412 -2.2402180 2.5638089 5.2854858 -1.8771128 -1.7008676 -0.1921968 0.7975146 -2.7284883 0.2457999 -0.9303121 -1.876257 -2.2321652 -2.0429843
M -0.6047750 3.678627 0.3775934 -0.1045283 -0.0638672 -1.8135474 -0.5729297 -0.2212046 -0.4385743 -0.5618393 0.2943020 -0.8096507 -0.8409468 0.8951194 -0.2593801 -0.1115049 -0.0464961 -0.6202754 -0.0042126 0.139225 -0.1782898 -1.8854336 -0.6997346 -0.5084394 -0.7968143 -0.8960554 -0.7939243 -2.2478484 0.0907235 0.5927939 0.1135766 0.0839364 0.4600360 0.2396026 -1.147825 0.9542984 0.3887704 
#> $`200fmol-50fmol`
NUD4B_HUMAN (+1) A0A7P0T808_HUMAN (+1) A0A8I5KU53_HUMAN (+1) ZN840_HUMAN CC85C_HUMAN C9JEV0_HUMAN (+1) C9JNU9_HUMAN CYC_BOVIN TRFE_BOVIN F8W0H2_HUMAN H0Y7V7_HUMAN (+1) H0YD14_HUMAN H3BUF6_HUMAN H7C1W4_HUMAN (+1) H7C3M7_HUMAN TLR3_HUMAN LRIG2_HUMAN RAB3D_HUMAN ADH1_YEAST LYSC_CHICK BGAL_ECOLI CYTA_HUMAN KPCB_HUMAN LIPL_HUMAN CO6_HUMAN BGAL_HUMAN SYTC_HUMAN CASPE_HUMAN DCAF6_HUMAN DALD3_HUMAN HGNAT_HUMAN RFFL_HUMAN RN185_HUMAN ZN462_HUMAN ALKB7_HUMAN POLK_HUMAN ACAD8_HUMAN
200fmol mean 1.2129798 0.9476617 1.6646748 -0.8707864 -0.9704449 -2.4355408 -1.0109247 4.803646 4.9771724 -0.1293629 0.8123346 0.9012464 -1.0788546 -1.6920949 0.8326191 -1.8320831 -0.7943162 0.7906959 6.301899 5.179787 6.103709 -0.3601495 0.8966850 -0.7324823 -2.425913 2.4717449 5.4536305 -3.2541540 -1.7246925 0.7037732 1.1711891 -2.3300528 -0.0848047 -0.4918852 -1.5365371 -2.0645198 -1.4177061
50fmol mean 1.5018335 -1.4867645 0.7414668 -1.2046036 -0.6978038 -0.7687851 -0.7991065 3.776136 4.5984741 0.1523311 0.2432124 1.2303125 -0.7489724 -2.6300921 0.6353621 -1.8284465 -0.9168044 1.2127423 5.071370 3.791556 4.984805 1.2366567 1.1905031 -0.7024215 -1.841811 3.0118366 5.6824480 -0.7531886 -1.7462293 -0.4885937 0.7407263 -2.7704565 0.0157819 -1.0501134 -1.3023443 -2.7093144 -2.2373695
A 1.3574066 -0.2695514 1.2030708 -1.0376950 -0.8341244 -1.6021630 -0.9050156 4.289891 4.7878233 0.0114841 0.5277735 1.0657795 -0.9139135 -2.1610935 0.7339906 -1.8302648 -0.8555603 1.0017191 5.686634 4.485671 5.544257 0.4382536 1.0435940 -0.7174519 -2.133862 2.7417908 5.5680392 -2.0036713 -1.7354609 0.1075898 0.9559577 -2.5502547 -0.0345114 -0.7709993 -1.4194407 -2.3869171 -1.8275378
M -0.2888537 2.4344262 0.9232081 0.3338172 -0.2726410 -1.6667557 -0.2118181 1.027510 0.3786983 -0.2816940 0.5691222 -0.3290661 -0.3298822 0.9379972 0.1972570 -0.0036365 0.1224882 -0.4220464 1.230530 1.388231 1.118905 -1.5968062 -0.2938181 -0.0300608 -0.584102 -0.5400917 -0.2288175 -2.5009654 0.0215368 1.1923669 0.4304628 0.4404038 -0.1005866 0.5582282 -0.2341928 0.6447946 0.8196634

Details

The result of method = "MA" is to generate the data for an MA plot, which plots the average fold change between two conditions (y-axis) against the average abundance of that protein (x-axis). This is helpful for evaluating whether a fold-change difference is being enhanced by low overall intensities (e.g. a change from 200 to 400 is the same fold-change as from 20,000 to 40,000, but the latter is a more robust measurement and less susceptible to noise interference).

PCA

Example

In the case of dataImput, one protein, namely “ALBU_BOVIN”, has constant values, leading to the error message. We choose to remove this protein in the principal component analysis (PCA).

names(dataImput)[sapply(dataImput, function(col) length(unique(col)) == 1)]
#> character(0)
dataPCA <- dataImput[, colnames(dataImput) != "ALBU_BOVIN"]
anlys_pca <- analyze.pca(dataPCA, center = TRUE, scale = TRUE)
PC1 PC2 PC3 PC4 PC5 PC6 PC7 PC8 PC9 PC10 PC11 PC12
NUD4B_HUMAN (+1) 0.2661771 -0.0113706 0.0552567 0.0083972 -0.1097609 -0.0491181 -0.0452318 -0.1702756 0.1244288 0.1282357 0.0564716 0.2065817
A0A7P0T808_HUMAN (+1) -0.2172426 -0.0775847 -0.1172228 0.2626137 -0.0485316 0.0502136 -0.1289626 0.0414284 0.1131366 0.2687172 0.0022229 0.2073718
A0A8I5KU53_HUMAN (+1) 0.0214544 -0.1948698 0.3220138 -0.0651332 0.1243099 -0.1787447 0.1395030 -0.0577622 -0.1939864 0.2064187 0.2609325 0.0252999
ZN840_HUMAN 0.0812586 -0.1527426 -0.0831556 0.1261251 -0.3674532 0.0887771 0.1614088 0.4020190 -0.2770179 -0.1656343 -0.1712628 0.1483799
CC85C_HUMAN 0.0470026 0.0829721 0.1168027 -0.3679799 -0.3201675 0.3069515 -0.1499944 -0.0483616 -0.0255806 -0.0622663 0.1901357 -0.0199917
C9JEV0_HUMAN (+1) 0.1731910 0.2012503 -0.1808202 -0.0639492 0.0043209 -0.1637345 -0.1193336 0.1356772 0.1899883 -0.0644367 -0.0609587 0.0554146
C9JNU9_HUMAN 0.1663077 -0.0270762 0.2682660 -0.0485203 -0.0162885 0.3281013 0.0794702 -0.1090887 -0.0646373 -0.1739993 -0.4179239 0.0390796
CYC_BOVIN 0.1135607 -0.2709043 -0.1789681 0.0062931 0.0477755 0.0965189 -0.0612769 0.0437122 -0.2244575 0.0143616 0.0538525 -0.2268740
TRFE_BOVIN 0.1873490 -0.1949425 -0.0336901 -0.1867213 0.0812859 0.1715941 -0.0534990 -0.0846296 0.0203016 0.1043512 -0.2755911 0.3304031
F8W0H2_HUMAN 0.2442243 0.0520695 0.0272016 -0.0330347 -0.0823934 0.1971741 -0.0550810 -0.2656510 -0.1773933 0.1230071 0.1367055 0.0622682
H0Y7V7_HUMAN (+1) -0.0504906 -0.2009404 -0.2906138 -0.0567632 -0.1451508 0.2159733 0.2238544 -0.1005611 0.2101825 0.0328399 0.0771365 -0.0663545
H0YD14_HUMAN 0.2626802 0.0284817 -0.0660155 0.0044856 0.0072984 0.1464363 0.0063931 0.0587555 0.1938316 -0.0485988 0.2459487 0.1278853
H3BUF6_HUMAN 0.2554844 0.0231902 0.0617154 0.0565942 0.1078012 0.0291533 -0.0158560 0.3264118 0.0281075 0.0378134 0.0361614 -0.0729308
H7C1W4_HUMAN (+1) -0.0999928 -0.1102991 -0.2739694 -0.1997307 -0.1484560 -0.2660805 -0.2406075 -0.2079829 -0.1443291 0.0094168 -0.1492226 -0.1482792
H7C3M7_HUMAN 0.1416175 -0.1791373 0.1614227 0.0737481 0.2066780 -0.1952269 -0.1933277 0.0080118 0.0118496 -0.1042807 -0.4930396 -0.0518115
TLR3_HUMAN 0.1299743 -0.0880416 -0.1347556 0.3224950 -0.2996605 -0.0334661 0.1111681 -0.3230490 0.0125494 0.0676457 -0.1048457 -0.1427992
LRIG2_HUMAN 0.0677596 -0.0893051 -0.1714711 -0.2517146 -0.4336555 -0.0523802 0.1497782 0.2376007 0.1693182 0.1148453 -0.0926185 0.0182974
RAB3D_HUMAN 0.2605899 0.0956436 0.0512783 0.0235534 -0.0566763 -0.0448168 -0.0809970 0.0900338 -0.1019880 -0.0584664 0.1718300 0.1267921
ADH1_YEAST 0.0711538 -0.2939140 -0.0663570 -0.0419019 0.1797889 0.1408954 -0.1267386 0.0773349 -0.1075854 -0.0898791 0.1615757 0.0244164
LYSC_CHICK 0.0490580 -0.3102726 -0.1309799 -0.0889801 0.0698650 0.1174480 -0.0113325 0.0068341 -0.1086235 0.0096390 0.0419618 -0.3387204
BGAL_ECOLI 0.1021770 -0.2821178 -0.0923051 -0.0808380 0.1171326 0.1258499 -0.1864386 0.0810218 -0.0600613 0.1078369 -0.0096848 0.2943592
CYTA_HUMAN 0.1825842 0.2188534 -0.1410898 -0.0305980 0.0945851 0.0058770 -0.1488556 0.1176593 0.0901028 -0.0849242 0.0922081 0.0794799
KPCB_HUMAN 0.2405626 0.0143265 -0.0507555 0.2414554 0.0163538 0.0598792 -0.1319232 -0.2157990 0.0302455 0.0920355 -0.0914497 -0.1086099
LIPL_HUMAN 0.2212831 -0.0973339 -0.0526197 -0.0719130 -0.1306337 -0.2878583 -0.0852227 -0.0425835 -0.0187239 0.2808939 -0.0699726 0.0820620
CO6_HUMAN 0.2137856 0.1161407 -0.0474913 0.0159156 -0.0765717 -0.2504026 0.2008667 0.2062349 -0.2339632 0.1350248 0.0262576 -0.1321902
BGAL_HUMAN 0.2635019 0.0815396 0.0321302 -0.0060343 0.0408999 -0.0529020 0.1304704 0.0040830 -0.1414278 -0.1222280 0.0860472 -0.1240084
SYTC_HUMAN 0.2731386 -0.0301576 0.0583046 0.0600839 0.0421041 0.0809829 -0.0415732 0.0708467 0.0603962 0.0358380 0.0384621 -0.4933504
CASPE_HUMAN 0.1595510 0.2484782 -0.1297485 0.0558357 -0.0209519 -0.0623907 -0.0148283 0.0733778 0.1723267 -0.1165495 -0.2083018 -0.0350683
DCAF6_HUMAN 0.0423154 -0.0641535 0.1689487 0.4157113 -0.2029096 0.0364136 -0.3875185 0.1061854 -0.0801830 -0.0087515 0.1322281 0.1234403
DALD3_HUMAN -0.0802794 -0.2666928 -0.0020154 -0.0510110 0.1365553 0.0199969 -0.0753417 0.3693074 0.3369189 0.0516420 -0.0388980 -0.1099699
HGNAT_HUMAN 0.0893249 -0.2229774 0.1535600 0.0257680 -0.0938518 -0.1579944 0.3662152 -0.1714206 0.1567205 -0.2798585 -0.0055094 0.1250200
RFFL_HUMAN 0.0614846 -0.1466767 -0.1130046 -0.2605200 -0.0177962 -0.4087473 -0.2506520 -0.1139699 -0.0617605 -0.3220286 0.1553084 0.1032499
RN185_HUMAN -0.0812159 0.0214999 0.3559595 -0.2297742 -0.2163234 0.0506643 -0.2062914 0.0311372 0.0534869 -0.2774764 -0.0439538 -0.1452909
ZN462_HUMAN 0.0437786 -0.1833759 0.3369196 -0.0263300 -0.0312656 -0.2179673 0.2330185 0.1225747 0.0376236 0.1752589 0.0446872 0.0932921
ALKB7_HUMAN 0.2218357 -0.0238417 -0.0515299 -0.1425276 0.2550007 0.0144762 0.1661079 -0.1488516 0.3315438 0.0183069 0.0636078 0.0119593
POLK_HUMAN 0.0092264 -0.1803050 0.2568307 0.1126046 -0.2618902 -0.0833389 -0.2315772 -0.0447542 0.4250459 0.0676487 0.1160992 -0.1885445
ACAD8_HUMAN 0.0058755 -0.2078893 -0.1544895 0.3085465 0.0284110 -0.0477976 0.0901488 -0.0415870 0.0615727 -0.5290980 0.1945109 0.1057147

Details

PCA is a powerful technique used in data analysis to simplify and reduce the dimensionality of large datasets. It transforms original variables into uncorrelated components that capture the maximum variance. By selecting a subset of these components, PCA projects the data points onto these key directions, enabling visualization and analysis in a lower-dimensional space. This aids in identifying patterns and relationships within complex datasets.

For PCA, the arguments center and scale are used to center the data to zero mean and scale to unit variance, with default setting at TRUE.

Note: Data scaling is done to ensure that the scale differences between different features do not affect the results of PCA. If not scaled, features with larger scales will dominate the computation of principal components (PCs).
Note: The most common error message for the PCA is “Cannot rescale a constant/zero column to unit variance.” This clearly occurs when columns representing proteins contain only zeros or have constant values. Typically, there are two ways to address this error: one is to remove these proteins, and the other is to set scale = FALSE.

PLS-DA

Example 

anlys_plsda <- analyze.plsda(dataImput, method = "kernelpls",
                             center = TRUE, scale = FALSE)
Comp 1 Comp 2 Comp 3 Comp 4 Comp 5 Comp 6 Comp 7 Comp 8 Comp 9 Comp 10 Comp 11
NUD4B_HUMAN (+1) -0.0878451 -0.1011609 0.0435591 -0.0223030 0.1343446 -0.0793665 -0.0704195 -0.0080244 0.0556818 0.2019409 0.0904704
A0A7P0T808_HUMAN (+1) 0.6555683 0.4283293 -0.4618501 -0.2260494 0.4007311 -0.2248149 -0.1373728 0.0905163 -0.0119397 0.1480421 -0.0078783
A0A8I5KU53_HUMAN (+1) 0.1015665 -0.3093573 0.4244665 -0.2918281 -0.1675887 -0.0876485 0.1768349 0.4570475 -0.0874848 0.0029154 0.2508353
ZN840_HUMAN 0.0150602 -0.1237765 -0.0287893 0.0185583 0.1899102 -0.1681087 0.0650244 -0.4492453 0.4776570 -0.3857429 -0.0876550
CC85C_HUMAN -0.0569953 0.0154783 0.1049973 0.1832721 -0.1717661 0.0198347 -0.3672053 -0.2468678 0.3515094 0.1364327 -0.0265155
C9JEV0_HUMAN (+1) -0.3141895 0.0742554 -0.1924158 0.1798450 0.1895171 -0.3543465 -0.0988269 0.2795523 -0.1595377 -0.0761303 -0.1428517
C9JNU9_HUMAN -0.0779428 -0.1167255 0.1393598 -0.1605541 -0.0419710 0.1821040 -0.1165818 -0.5479751 0.0016749 0.1305581 0.0267092
CYC_BOVIN 0.0355414 -0.2951654 -0.1578367 0.0291097 0.0587003 -0.0479550 -0.0124603 -0.0914774 0.1284605 -0.0380388 0.1275646
TRFE_BOVIN -0.0317380 -0.1986103 -0.0299751 0.0032207 -0.0393434 -0.0502623 -0.0877941 -0.1698850 -0.0381481 0.2055974 -0.0384345
F8W0H2_HUMAN -0.0840781 -0.0649539 0.0062435 -0.0340408 0.0261689 0.0235405 -0.0840596 -0.0939826 0.1548745 0.2046532 0.1925352
H0Y7V7_HUMAN (+1) 0.0599377 -0.0700030 -0.1054440 0.1203902 0.0324412 0.0805284 0.0433883 -0.0676874 0.2130588 0.1778651 -0.2261512
H0YD14_HUMAN -0.1146096 -0.1035337 -0.0447822 -0.0559923 0.1138489 0.0604790 -0.1435081 0.0206761 0.1263967 0.0782525 -0.0480208
H3BUF6_HUMAN -0.1041022 -0.1007199 -0.0021147 -0.1739466 0.0773905 -0.0791580 -0.1010617 0.0122957 0.0108227 -0.1262139 -0.0202390
H7C1W4_HUMAN (+1) 0.1196957 -0.0317759 -0.1630352 0.6662041 -0.0643975 -0.4341280 0.1488242 0.1057695 -0.1200804 0.1299113 0.2053694
H7C3M7_HUMAN -0.0101254 -0.1338164 0.0506149 -0.0769448 0.0688441 -0.0905647 0.0010059 -0.0821184 -0.3466046 -0.0632792 0.0957705
TLR3_HUMAN -0.0070652 -0.0431710 -0.0245317 0.0100383 0.2078266 -0.0168581 0.0920238 -0.1218879 0.1328421 0.1621621 0.1092826
LRIG2_HUMAN -0.0115352 -0.0660558 -0.0080433 0.2086486 0.0499283 -0.2710415 -0.0142559 -0.1057668 0.3332981 0.0365818 -0.3149645
RAB3D_HUMAN -0.0965299 -0.0475043 0.0210609 -0.0420115 0.0772828 -0.0527535 -0.0686250 0.0099140 0.0732519 -0.0504485 0.1170628
ADH1_YEAST 0.0663190 -0.3121837 -0.1030544 -0.0187514 -0.0315635 0.1084271 -0.1556204 0.0018891 -0.0618222 -0.1225742 0.0599116
LYSC_CHICK 0.0896312 -0.3222697 -0.1209861 0.0785042 -0.0347675 0.0136149 -0.0259660 -0.0761407 0.0764810 0.0367909 -0.0370304
BGAL_ECOLI 0.0451615 -0.3097967 -0.1153555 -0.0062625 -0.0210757 -0.1027940 -0.1989268 -0.0636903 -0.0509210 0.0377599 0.0329697
CYTA_HUMAN -0.3136960 0.0677818 -0.2106804 -0.0093124 0.0790596 -0.0526301 -0.2498384 0.2270980 -0.0761540 -0.0890951 0.0557064
KPCB_HUMAN -0.0850475 -0.0786471 -0.0528652 -0.1117253 0.2012475 0.0000287 -0.0481383 -0.0944865 -0.0282465 0.2025368 0.2327567
LIPL_HUMAN -0.0606082 -0.1392547 0.0164007 0.0621910 0.1216315 -0.3210855 0.0219300 0.0810506 0.0294751 0.1359454 0.0924105
CO6_HUMAN -0.1172698 -0.0333863 0.0050761 -0.0499494 0.0829277 -0.2326305 0.1707073 0.0998001 0.1768494 -0.1090753 0.0492317
BGAL_HUMAN -0.1354855 -0.0850057 0.0244161 -0.0657709 0.0654603 0.0133423 0.0607371 0.0156065 0.0757379 -0.0309448 0.0998194
SYTC_HUMAN -0.0980984 -0.1350389 0.0076404 -0.1315205 0.1147707 -0.0231823 -0.1170825 -0.0502408 0.0292206 0.0455124 0.0535278
CASPE_HUMAN -0.3938587 0.2108454 -0.1942282 0.0087377 0.3256512 -0.1623941 0.0271134 -0.0870876 -0.1255307 -0.0121095 -0.1378310
DCAF6_HUMAN 0.0268975 -0.0088578 0.0476034 -0.1382185 0.2408494 -0.0506598 -0.2238316 -0.1445788 0.0308343 -0.1598962 0.3168399
DALD3_HUMAN 0.1445632 -0.1640072 -0.0359533 -0.0455564 -0.0232551 -0.0950414 -0.1921970 0.1038114 -0.2317482 -0.2123031 -0.4695513
HGNAT_HUMAN 0.0247228 -0.2312723 0.2079354 0.0446899 0.2099683 0.1787024 0.2818864 -0.0626436 0.0192533 0.0497766 -0.1913015
RFFL_HUMAN -0.0065966 -0.1667078 0.0137115 0.4315663 0.0123620 -0.1752065 0.0191264 0.3026119 -0.2436960 -0.1649764 0.1508422
RN185_HUMAN 0.0224085 0.0510364 0.2325357 0.1088302 -0.1210609 0.0248317 -0.2731382 -0.2080871 -0.0436765 -0.1309695 -0.0191369
ZN462_HUMAN 0.0735380 -0.2755487 0.4795177 -0.2856937 0.0581440 -0.2910819 0.1764994 0.1974836 0.0015812 -0.1068532 -0.2283853
ALKB7_HUMAN -0.1534580 -0.2141033 -0.0426737 -0.0596206 0.0094284 0.1585526 0.0295445 0.2608349 -0.1898564 0.4014002 -0.2688850
POLK_HUMAN 0.1406435 -0.1578093 0.3810856 -0.0048664 0.4837772 -0.1983768 -0.5926675 -0.0719198 -0.1089691 0.2168022 -0.0407733
ACAD8_HUMAN 0.0895749 -0.1621447 -0.1240768 0.0671191 0.3837676 0.4192277 0.2065079 0.0022848 -0.0630324 -0.3461769 0.0403876

Details

Partial least squares-discriminant analysis (PLS-DA) adapts PLS regression for supervised classification. Rather than simply finding directions of maximal variances in the predictors as PCA does, PLS-DA extracts latent components that maximize the covariance between predictors and dummy-coded group labels. This ensures that the resulting components optimally separate predefined groups and yields variable-importance scores directly tied to classification.

For PLS-DA, the argument method specifies which multivariate regression algorithm to use:

  1. “kernelpls”: Kernel algorithm (Dayal and MacGregor 1997).

  2. “widekernelpls”: Wide kernel algorithm (Rännar et al. 1994).

  3. “simpls”: SIMPLS algorithm (Jong 1993).

  4. “oscorespls”: NIPALS algorithm (classical orthogonal scores algorithm) (Martens and Næs 1989).

The argument ncomp sets the number of components to include in the model. It defaults to min(n-1, p). The arguments center and scale control whether the data are centered to zero mean and scaled to unit variance, respectively.

Reference

← Previous

Next →

Benjamini, Yoav, and Yosef Hochberg. 1995. “Controlling the False Discovery Rate: A Practical and Powerful Approach to Multiple Testing.” Journal of the Royal Statistical Society: Series B (Methodological) 57 (1): 289–300. https://doi.org/10.1111/j.2517-6161.1995.tb02031.x.
Benjamini, Yoav, and Daniel Yekutieli. 2001. “The Control of the False Discovery Rate in Multiple Testing Under Dependency.” The Annals of Statistics 29 (4): 1165–88. https://doi.org/10.1214/aos/1013699998.
Bonferroni, C. E. 1936. “Teoria Statistica Delle Classi e Calcolo Delle Probabilità.” Pubblicazioni Del R Istituto Superiore Di Scienze Economiche e Commerciali Di Firenze 8: 3–62.
Dayal, Bhupinder S., and John F. MacGregor. 1997. “Improved PLS Algorithms.” Journal of Chemometrics 11 (1): 73–85. https://doi.org/10.1002/(SICI)1099-128X(199701)11:1<73::AID-CEM435>3.0.CO;2-\%23.
Hochberg, Yosef. 1988. “A Sharper Bonferroni Procedure for Multiple Tests of Significance.” Biometrika 75 (4): 800–802. https://doi.org/10.1093/biomet/75.4.800.
Holm, Sture. 1979. “A Simple Sequentially Rejective Multiple Test Procedure.” Scandinavian Journal of Statistics 6 (2): 65–70. https://www.jstor.org/stable/4615733.
Hommel, Gerhard. 1988. “A Stagewise Rejective Multiple Test Procedure Based on a Modified Bonferroni Test.” Biometrika 75 (2): 383–86. https://doi.org/10.1093/biomet/75.2.383.
Jong, Sijmen de. 1993. “SIMPLS: An Alternative Approach to Partial Least Squares Regression.” Chemometrics and Intelligent Laboratory Systems 18 (3): 251–63. https://doi.org/10.1016/0169-7439(93)85002-X.
Martens, Harald, and Tormod Næs. 1989. Multivariate Calibration. Chichester.
Rännar, Stefan, Fredrik Lindgren, Paul Geladi, and Svante Wold. 1994. “A PLS Kernel Algorithm for Data Sets with Many Variables and Fewer Objects. Part 1: Theory and Algorithm.” Journal of Chemometrics 8 (2): 111–25. https://doi.org/10.1002/cem.1180080204.
Smyth, Gordon K. 2004. “Linear Models and Empirical Bayes Methods for Assessing Differential Expression in Microarray Experiments.” Statistical Applications in Genetics and Molecular Biology 3 (1). https://doi.org/10.2202/1544-6115.1027.