options(java.parameters = "-Xmx4000M")
library(SELEX)
library(SelexGLM)
library(grid)
workDir = "./cache/"
selex.config(workingDir=workDir, maxThreadNumber=4)
### LOCAL PATHS NEED TO BE RE-DEFINED TO RUN OFF OF MY COMPUTER
##################################################################
selexDir = "/Users/gabriella/Columbia/SELEX/"
#rawdataDir = "/Users/gabriella/Columbia/rawdata/Pufall/"
processedDataDir = "/Users/gabriella/Columbia/SplitFastqData/Pufall/ConcatFiles/"
# CLUSTER VERSIONS ARE COMMENTED OUT
#selexDir = "/vega/hblab/users/gdm2120/SELEX/SELEX/"
#rawdataDir = "/vega/hblab/projects/selex/rawdata/Pufall"
#processedDataDir = "/vega/hblab/users/gdm2120/SplitFastqData/Pufall/"
##################################################################
saveDir = "gabriella/SelexGLMtest/ShapeFixedValuesSymmetry"
dir.create(file.path(selexDir, saveDir), showWarnings = FALSE, recursive = TRUE)
shapeTable = read.table(paste(selexDir, "gabriella/ShapeParamData/ShapeTableOrthogonal.txt", sep = ""), sep = "\t",
stringsAsFactors = FALSE)
ST = shapeTable[,c(1, 14:19)]
colnames(ST) = c("Sequence", "MGW", "ProT", "HelTA",
"HelTB", "RollA", "RollB")
selex.defineSample('r0.Pufall',
paste(processedDataDir, "/Demultiplexed.R0.fastq.gz", sep = ""),
'r0',
0, 23, '', 'TGGAA')
selex.defineSample('AR.R8',
paste(processedDataDir,"/AR.R8.fastq.gz",sep = ""),
'AR-DBD',
8, 23, '', 'TGGAA')
selex.defineSample('AR.R7',
paste(processedDataDir,"/AR.R7.fastq.gz",sep = ""),
'AR-DBD',
7, 23, '', 'TGGAA')
r0 = selex.sample(seqName = 'r0.Pufall', sampleName='r0', round = 0)
r0.split = selex.split(r0)
r0.train = r0.split$train
r0.test = r0.split$test
dataSample = selex.sample(seqName = 'AR.R8', sampleName = 'AR-DBD', round = 8)
# MARKOV MODEL BUILT
kmax = selex.kmax(sample = r0.test)
mm = selex.mm(sample = r0.train, order = NA, crossValidationSample =r0.test, Kmax = kmax, mmMethod = "TRANSITION")
mmscores = selex.mmSummary(sample = r0.train)
ido = which(mmscores$R==max(mmscores$R))
mm.order = mmscores$Order[ido]
libLen = as.numeric(as.character(selex.getAttributes(dataSample)$VariableRegionLength))
kLen = 15
#data.probeCounts = getProbeCounts(dataSample, markovModel = mm)
#save(data.probeCounts, file = paste(selexDir, saveDir, "/data.probeCounts.RData", sep = ""))
load(file = paste(selexDir, saveDir, "/data.probeCounts.RData", sep = ""))
#data.kmerTable = getKmerCountAffinities(dataSample, k = kLen, minCount = 100, markovModel = mm)
#save(data.kmerTable, file = paste(selexDir, saveDir, "/data.kmerTable.RData", sep = ""))
load(file = paste(selexDir, saveDir, "/data.kmerTable.RData", sep = ""))
# Inputs about library are data specific
load(paste(selexDir, "/gabriella/SelexGLMtest/ShapeSymmetry/model.RData", sep = ""))
Shape.values = ModelTest@features@Shape@Shape.values[c("Shape.MGW", "Shape.HelTA", "Shape.HelTB"),]
ModelTest = model(name = "AR-DBD R8 Nucleotides + Fixed Shape Values (Rev. Comp. Sym.)",
varRegLen = libLen,
leftFixedSeq = "GTTCAGAGTTCTACAGTCCGACGATC",
rightFixedSeq ="TGGAATTCTCGGGTGCCAAGG",
consensusSeq = "RGWACANNNTGTWCY",
affinityType = "AffinitySym",
leftFixedSeqOverlap = 5,
minAffinity = 0.01,
missingValueSuppression = .5,
minSeedValue = .01,
upFootprintExtend = 4,
confidenceLevel = .99,
rounds = list(c(8)),
rcSymmetric = TRUE,
verbose = FALSE,
includeShape = TRUE,
shapeTable = ST,
shapeParams = list(c("MGW", "HelT")),
Shape.values = Shape.values,
useFixedValuesOffset.Shape = TRUE,
Shape.set = c(0))
getFeatureDesign(ModelTest)
## Feature design for object of class 'model'
##
## seedLen: 15
## upFootprintExtend: 4
## downFootprintExtend: 4
## rcSymmetric: TRUE
##
## Slot "N":
## N.upFootprintExtend: 4
## N.downFootprintExtend: 4
## N.set: 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23
## Number of previous iterations: 0
##
## Slot "Intercept":
## Number of Views per Strand of DNA: 11
## Number of Rounds: 1 (8)
## Number of previous iterations: 0
##
## Slot "Shape":
## ShapeParamsUsed: HelT MGW
## Shape.upFootprintExtend: 0
## Shape.downFootprintExtend: 0
## Shape.set: 0
## Number of previous iterations: 0
# Add seed model
addSeedPsam(ModelTest) = seedTable2psam(ModelTest, data.kmerTable)
# Model nucleotide Betas after seed PSAM is added
print(getValues(getN(ModelTest)))
## 1 2 3 4 5 6 7 8 9
## N.A 0 0 0 0 0.00000000 -1.2968295 -0.03073087 0.000000 -1.296829
## N.C 0 0 0 0 -0.60728754 -1.2968295 -0.25628921 -1.296829 0.000000
## N.G 0 0 0 0 -0.09864725 0.0000000 -0.34036727 -1.296829 -1.296829
## N.T 0 0 0 0 -0.42644057 -0.5591611 0.00000000 -1.296829 -1.296829
## 10 11 12 13 14 15
## N.A 0.0000000 -0.40799975 -0.1359377 -0.09020211 -0.7968295 -1.296829
## N.C -1.2968295 0.00000000 0.0000000 -0.36546623 -0.3957275 -1.296829
## N.G -0.3957275 -0.36546623 0.0000000 0.00000000 -1.2968295 0.000000
## N.T -0.7968295 -0.09020211 -0.1359377 -0.40799975 0.0000000 -1.296829
## 16 17 18 19 20 21 22 23
## N.A -1.296829 0.00000000 -0.5591611 -0.42644057 0 0 0 0
## N.C -1.296829 -0.34036727 0.0000000 -0.09864725 0 0 0 0
## N.G -1.296829 -0.25628921 -1.2968295 -0.60728754 0 0 0 0
## N.T 0.000000 -0.03073087 -1.2968295 0.00000000 0 0 0 0
plot(ModelTest@features@N, Ntitle = "AR-DBD R8 Nucleotides+Fixed Shape Values\nSeeding Model", ddG = TRUE)
Next we score the probes using topModelMatch
sample1 = sample(nrow(data.probeCounts), 1000000)
data = data.probeCounts[sample1,]
#data = data.probeCounts
data = topModelMatch(data, ModelTest)
# Uses aligned probes to build design matrix
data = addDesignMatrix(data, ModelTest)
designMatrixSummary = getDesignMatrix(ModelTest, data)
## No shape parameters included in fit.
print("Round summary: ")
## [1] "Round summary: "
print (designMatrixSummary$Round)
## 8 Total
## Round 998602 998602
print("View/strand orientation summary: ")
## [1] "View/strand orientation summary: "
print (designMatrixSummary$Intercept)
## View.1 View.2 View.3 View.4 View.5 View.6 View.7 View.8 View.9
## Strand.F 5057 64448 97022 69913 107574 129495 141334 153755 110908
## Strand.R 0 0 0 0 0 0 0 0 0
## View.10 View.11 StrandTotal
## Strand.F 56913 62183 998602
## Strand.R 0 0 0
print("Mono-nucleotide summary: ")
## [1] "Mono-nucleotide summary: "
print (designMatrixSummary$N)
## N.A N.C N.G N.T
## 1 583012 637923 348200 428069
## 2 718433 470565 371671 436535
## 3 908581 499325 171102 418196
## 4 926804 254270 407373 408757
## 5 1289516 18250 647348 42090
## 6 638 213 1981071 15282
## 7 843219 104875 93365 955745
## 8 1994876 388 1103 837
## 9 684 1995725 221 574
## 10 1878970 677 114640 2917
## 11 80186 1112788 111512 692718
## 12 342880 655722 0 0
# # Constructs regression expression with independent features using design matrix
regressionFormula = updatedRegressionFormula(data, ModelTest)
print("Regression Formula: ")
## [1] "Regression Formula: "
print (regressionFormula)
## [1] "ObservedCount ~ offset(logProb)+offset(fixedSddG)+N.A1+N.G1+N.T1+N.C2+N.G2+N.T2+N.C3+N.G3+N.T3+N.C4+N.G4+N.T4+N.C5+N.G5+N.T5+N.A6+N.C6+N.T6+N.A7+N.C7+N.G7+N.C8+N.G8+N.T8+N.A9+N.G9+N.T9+N.C10+N.G10+N.T10+N.A11+N.G11+N.T11+N.A12"
fit = glm(regressionFormula,
data=data,
family = poisson(link="log"))
## Warning: glm.fit: fitted rates numerically 0 occurred
## Warning: glm.fit: algorithm did not converge
## Warning: glm.fit: fitted rates numerically 0 occurred
## Warning in glm(regressionFormula, data = data, family = poisson(link
## = "log")): fitting to calculate the null deviance did not converge --
## increase 'maxit'?
summary(fit)
##
## Call:
## glm(formula = regressionFormula, family = poisson(link = "log"),
## data = data)
##
## Deviance Residuals:
## Min 1Q Median 3Q Max
## -56.041 -1.210 -0.390 0.918 67.797
##
## Coefficients:
## Estimate Std. Error z value Pr(>|z|)
## (Intercept) 39.3616830 0.0026175 15037.654 <2e-16 ***
## N.A1 0.0061139 0.0001378 44.380 <2e-16 ***
## N.G1 0.0072465 0.0001705 42.499 <2e-16 ***
## N.T1 -0.0362859 0.0001706 -212.706 <2e-16 ***
## N.C2 -0.0529322 0.0001560 -339.385 <2e-16 ***
## N.G2 0.0238781 0.0001452 164.408 <2e-16 ***
## N.T2 -0.0345194 0.0001754 -196.775 <2e-16 ***
## N.C3 -0.0550794 0.0001771 -310.934 <2e-16 ***
## N.G3 -0.1432979 0.0002368 -605.022 <2e-16 ***
## N.T3 -0.0393158 0.0001696 -231.836 <2e-16 ***
## N.C4 -0.1445761 0.0002581 -560.061 <2e-16 ***
## N.G4 -0.0623558 0.0001626 -383.492 <2e-16 ***
## N.T4 -0.1165680 0.0001654 -704.951 <2e-16 ***
## N.C5 -0.4237589 0.0010614 -399.240 <2e-16 ***
## N.G5 -0.0959986 0.0001477 -649.813 <2e-16 ***
## N.T5 -0.6078251 0.0006348 -957.519 <2e-16 ***
## N.A6 -0.9801375 0.0279512 -35.066 <2e-16 ***
## N.C6 -0.9163720 0.0625001 -14.662 <2e-16 ***
## N.T6 -0.7966607 0.0011788 -675.843 <2e-16 ***
## N.A7 -0.8070699 0.0001294 -6236.586 <2e-16 ***
## N.C7 -1.1118090 0.0002961 -3754.363 <2e-16 ***
## N.G7 -0.2800741 0.0003998 -700.572 <2e-16 ***
## N.C8 -2.2179610 29.6438029 -0.075 0.9404
## N.G8 -1.7297537 0.0195223 -88.604 <2e-16 ***
## N.T8 -0.6964109 0.0264497 -26.330 <2e-16 ***
## N.A9 -1.8600765 0.0188446 -98.706 <2e-16 ***
## N.G9 0.3026519 0.1254515 2.413 0.0158 *
## N.T9 -0.6568010 0.0475211 -13.821 <2e-16 ***
## N.C10 -1.0511035 0.0416668 -25.226 <2e-16 ***
## N.G10 -0.5121916 0.0003194 -1603.619 <2e-16 ***
## N.T10 -0.6164851 0.0050177 -122.863 <2e-16 ***
## N.A11 -0.3482993 0.0004190 -831.261 <2e-16 ***
## N.G11 -0.2860060 0.0003275 -873.373 <2e-16 ***
## N.T11 -0.0842063 0.0001299 -648.258 <2e-16 ***
## N.A12 -0.1058304 0.0001783 -593.436 <2e-16 ***
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## (Dispersion parameter for poisson family taken to be 1)
##
## Null deviance: 124521737 on 998601 degrees of freedom
## Residual deviance: 7149588 on 998567 degrees of freedom
## AIC: 8588943
##
## Number of Fisher Scoring iterations: 18
ModelTest = addNewBetas(ModelTest, data, fit)
## No shape parameters included in fit.
# # Nucleotide Features after first round of fitting
summary(ModelTest)
## An object of class 'model'
##
## Slot "name": AR-DBD R8 Nucleotides + Fixed Shape Values (Rev. Comp. Sym.)
## Slot "varRegLen": 23
## Slot "leftFixedSeq": GTTCAGAGTTCTACAGTCCGACGATC
## Slot "rightFixedSeq": TGGAATTCTCGGGTGCCAAGG
## Slot "leftFixedSeqOverlap": 5
## Slot "rightFixedSeqOverlap": 5
## Slot "confidenceLevel": 0.99
## Slot "minAffinity": 0.01
## Slot "missingValueSuppression": 0.5
## Slot "minSeedValue": 0.01
## Slot "seedLen": 15
## Slot "consensusSeq": [AG]G[AT]ACA[ACGT][ACGT][ACGT]TGT[AT]C[CT]
## Slot "upFootprintExtend": 4
## Slot "downFootprintExtend": 4
## Slot "fpLen": 23
##
## Fits a model of footprint length 23 for mono-nucleotide and shape features (shape = HelT, MGW) with fixed values for shape parameter positions not included in Shape.set used as offsets for the glm fit, and 11 view(s) per strand of DNA and 1 round(s) of data (round = 8) with reverse complement symmetry.
##
## Slot "regressionFormula": ObservedCount ~ offset(logProb)+offset(fixedSddG)+Round.8+N.A1+N.C1+N.G1+N.T1+N.A2+N.C2+N.G2+N.T2+N.A3+N.C3+N.G3+N.T3+N.A4+N.C4+N.G4+N.T4+N.A5+N.C5+N.G5+N.T5+N.A6+N.C6+N.G6+N.T6+N.A7+N.C7+N.G7+N.T7+N.A8+N.C8+N.G8+N.T8+N.A9+N.C9+N.G9+N.T9+N.A10+N.C10+N.G10+N.T10+N.A11+N.C11+N.G11+N.T11+N.A12+N.C12
##
## Slot "shapeParamsUsed[[1]]": HelT MGW
##
## Includes the following feature sub-classes:
## An object of class 'N'
## Fits 23 nucleotides for a feature model of length 23.
## Nucleotide features are reverse complement symmetric.
## Nucleotide beta values:
## 1 2 3 4 5
## N.A 0.006113935 0.00000000 0.00000000 0.00000000 0.00000000
## N.C 0.000000000 -0.05293223 -0.05507943 -0.14457613 -0.42375888
## N.G 0.007246529 0.02387814 -0.14329785 -0.06235576 -0.09599862
## N.T -0.036285917 -0.03451938 -0.03931584 -0.11656801 -0.60782509
## 6 7 8 9 10 11
## N.A -0.9801375 -0.8070699 0.0000000 -1.8600765 0.0000000 -0.34829933
## N.C -0.9163720 -1.1118090 -2.2179610 0.0000000 -1.0511035 0.00000000
## N.G 0.0000000 -0.2800741 -1.7297537 0.3026519 -0.5121916 -0.28600603
## N.T -0.7966607 0.0000000 -0.6964109 -0.6568010 -0.6164851 -0.08420628
## 12 13 14 15 16 17
## N.A -0.1058304 -0.08420628 -0.6164851 -0.6568010 -0.6964109 0.0000000
## N.C 0.0000000 -0.28600603 -0.5121916 0.3026519 -1.7297537 -0.2800741
## N.G 0.0000000 0.00000000 -1.0511035 0.0000000 -2.2179610 -1.1118090
## N.T -0.1058304 -0.34829933 0.0000000 -1.8600765 0.0000000 -0.8070699
## 18 19 20 21 22
## N.A -0.7966607 -0.60782509 -0.11656801 -0.03931584 -0.03451938
## N.C 0.0000000 -0.09599862 -0.06235576 -0.14329785 0.02387814
## N.G -0.9163720 -0.42375888 -0.14457613 -0.05507943 -0.05293223
## N.T -0.9801375 0.00000000 0.00000000 0.00000000 0.00000000
## 23
## N.A -0.036285917
## N.C 0.007246529
## N.G 0.000000000
## N.T 0.006113935
##
## Nucleotide beta errors:
## 1 2 3 4 5
## N.A 0.0001377629 0.0000000000 0.0000000000 0.0000000000 0.0000000000
## N.C 0.0000000000 0.0001559650 0.0001771418 0.0002581435 0.0010614148
## N.G 0.0001705119 0.0001452371 0.0002368472 0.0001625998 0.0001477328
## N.T 0.0001705920 0.0001754259 0.0001695850 0.0001653561 0.0006347914
## 6 7 8 9 10
## N.A 0.027951234 0.0001294089 0.00000000 0.01884457 0.0000000000
## N.C 0.062500090 0.0002961379 29.64380288 0.00000000 0.0416668313
## N.G 0.000000000 0.0003997790 0.01952225 0.12545152 0.0003193972
## N.T 0.001178766 0.0000000000 0.02644966 0.04752105 0.0050176758
## 11 12 13 14 15
## N.A 0.0004190012 0.000178335 0.0001298962 0.0050176758 0.04752105
## N.C 0.0000000000 0.000000000 0.0003274728 0.0003193972 0.12545152
## N.G 0.0003274728 0.000000000 0.0000000000 0.0416668313 0.00000000
## N.T 0.0001298962 0.000178335 0.0004190012 0.0000000000 0.01884457
## 16 17 18 19 20
## N.A 0.02644966 0.0000000000 0.001178766 0.0006347914 0.0001653561
## N.C 0.01952225 0.0003997790 0.000000000 0.0001477328 0.0001625998
## N.G 29.64380288 0.0002961379 0.062500090 0.0010614148 0.0002581435
## N.T 0.00000000 0.0001294089 0.027951234 0.0000000000 0.0000000000
## 21 22 23
## N.A 0.0001695850 0.0001754259 0.0001705920
## N.C 0.0002368472 0.0001452371 0.0001705119
## N.G 0.0001771418 0.0001559650 0.0000000000
## N.T 0.0000000000 0.0000000000 0.0001377629
##
##
## An object of class 'Intercept'
## Fits intercept(s) for 1 round(s) (round = 8).
## Intercept beta values:
## Round.8:
## [1] 39.36168
##
## Intercept beta errors:
## Round.8:
## [1] 0.002617542
##
##
##
## An object of class 'Shape'
## Fits 3 shape coefficients for 3 kinds of shape parameter(s) (shape = HelT, MGW) for a feature model of length 23.
## Shape features are reverse complement symmetric.
## Shape beta values:
## 1 2 3 4 5
## Shape.HelTA 0.012777702 0.009207855 0.01059514 0.02979068 -0.050599877
## Shape.HelTB 0.005339387 -0.003648404 -0.03570894 -0.01147808 -0.008140565
## Shape.MGW 0.011501160 -0.008920502 -0.02962055 -0.18939069 -0.087382882
## 6 7 8 9 10
## Shape.HelTA -0.04637238 -0.1065263 0.1929080 0.19583385 0.11049638
## Shape.HelTB -0.12908078 0.3710017 0.7437455 -0.03934332 -0.07502132
## Shape.MGW 0.06924589 0.3766392 1.2499150 0.17179767 -0.17550875
## 11 12 13 14 15
## Shape.HelTA 0.09218623 0.02171862 -0.03883268 -0.07502132 -0.03934332
## Shape.HelTB -0.03883268 0.02171862 0.09218623 0.11049638 0.19583385
## Shape.MGW -0.12381238 0.15027160 -0.12381238 -0.17550875 0.17179767
## 16 17 18 19 20
## Shape.HelTA 0.7437455 0.3710017 -0.12908078 -0.008140565 -0.01147808
## Shape.HelTB 0.1929080 -0.1065263 -0.04637238 -0.050599877 0.02979068
## Shape.MGW 1.2499150 0.3766392 0.06924589 -0.087382882 -0.18939069
## 21 22 23
## Shape.HelTA -0.03570894 -0.003648404 0.005339387
## Shape.HelTB 0.01059514 0.009207855 0.012777702
## Shape.MGW -0.02962055 -0.008920502 0.011501160
##
## Shape beta errors:
## 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23
## Shape.HelTA 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0
## Shape.HelTB 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0
## Shape.MGW 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0
vPheight = verticalPlot_height(ModelTest)
pM <- plot(ModelTest, plotTitle = "AR-DBD Nucleotide + Fixed Shape Values Fit", Nplot.ddG = TRUE, verticalPlots = TRUE)
ggplot2::ggsave(pM, file = paste(selexDir, saveDir, "/modelPlot.pdf", sep = ""), height = vPheight, width = 6)
ggplot2::ggsave(pM, file = paste(selexDir, saveDir, "/modelPlot.",1, ".pdf", sep = ""), height = vPheight, width = 6)
data = data.probeCounts[sample1,]
#data = data.probeCounts
data = topModelMatch(data, ModelTest)
data = addDesignMatrix(data, ModelTest)
if (nrow(data) > 0) {
designMatrixSummary.v2 = getDesignMatrix(ModelTest, data)
if ((all(designMatrixSummary.v2$N == designMatrixSummary$N)) & (all(designMatrixSummary.v2$Round == designMatrixSummary$Round)) & (all(designMatrixSummary.v2$Intercept == designMatrixSummary$Intercept))) {
print ("Stability Reached")
}
}
## No shape parameters included in fit.
for (i in 2:20) {
if (nrow(data) == 0) {
break
} else if ((all(designMatrixSummary.v2$N == designMatrixSummary$N)) & (all(designMatrixSummary.v2$Round == designMatrixSummary$Round)) & (all(designMatrixSummary.v2$Intercept == designMatrixSummary$Intercept))) {
break
}
data.nrow = nrow(data)
print (paste("i =",i))
designMatrixSummary = designMatrixSummary.v2
print("Round summary: ")
print (designMatrixSummary$Round)
print("Mono-nucleotide summary: ")
print (designMatrixSummary$N)
print("View/strand orientation summary: ")
print (designMatrixSummary$Intercept)
# # Constructs regression expression with independent features using design matrix
regressionFormula = updatedRegressionFormula(data, ModelTest)
print("Regression Formula: ")
print (regressionFormula)
fit = glm(regressionFormula,
data=data,
family = poisson(link="log"))
summary(fit)
ModelTest = addNewBetas(ModelTest, data, fit)
# # Nucleotide Features after first round of fitting
summary(ModelTest)
pM <- plot(ModelTest, plotTitle = "AR-DBD R8 Nucleotide+ Fixed Shape Values Fit", Nplot.ddG = TRUE, verticalPlots = TRUE)
ggplot2::ggsave(pM, file = paste(selexDir, saveDir, "/modelPlot.",i, ".pdf", sep = ""), height = vPheight, width = 6)
ggplot2::ggsave(pM, file = paste(selexDir, saveDir, "/modelPlot.pdf", sep = ""), height = vPheight, width = 6)
data = topModelMatch(data, ModelTest)
data = addDesignMatrix(data, ModelTest)
print(paste("Number of Observations in Design Matrix: ",nrow(data), sep = ""))
if (nrow(data) > 0) {
designMatrixSummary.v2 = getDesignMatrix(ModelTest, data)
if ((all(designMatrixSummary.v2$N == designMatrixSummary$N)) & (all(designMatrixSummary.v2$Round == designMatrixSummary$Round)) & (all(designMatrixSummary.v2$Intercept == designMatrixSummary$Intercept))) {
print (paste("Stability Reached after ", i, " iterations.", sep = ""))
break
}
} else {
print (paste("Algorithm failed to converge: No probes meet the confidence level requirement (Confidence Level:", ModelTest@confidenceLevel, ")", sep = ""))
}
}
## [1] "i = 2"
## [1] "Round summary: "
## 8 Total
## Round 998104 998104
## [1] "Mono-nucleotide summary: "
## N.A N.C N.G N.T
## 1 582528 637831 348184 427665
## 2 718056 470469 371677 436006
## 3 908425 498958 170857 417968
## 4 926424 254094 407115 408575
## 5 1289035 18433 646573 42167
## 6 720 334 1979935 15219
## 7 841699 104776 93505 956228
## 8 1993623 327 1099 1159
## 9 678 1994256 609 665
## 10 1877487 727 114882 3112
## 11 79687 1111962 111525 693034
## 12 342946 655158 0 0
## [1] "View/strand orientation summary: "
## View.1 View.2 View.3 View.4 View.5 View.6 View.7 View.8 View.9
## Strand.F 5091 64374 96871 69847 107518 129598 141257 153696 110851
## Strand.R 0 0 0 0 0 0 0 0 0
## View.10 View.11 StrandTotal
## Strand.F 56852 62149 998104
## Strand.R 0 0 0
## [1] "Regression Formula: "
## [1] "ObservedCount ~ offset(logProb)+offset(fixedSddG)+N.A1+N.G1+N.T1+N.C2+N.G2+N.T2+N.C3+N.G3+N.T3+N.C4+N.G4+N.T4+N.C5+N.G5+N.T5+N.A6+N.C6+N.T6+N.A7+N.C7+N.G7+N.C8+N.G8+N.T8+N.A9+N.G9+N.T9+N.C10+N.G10+N.T10+N.A11+N.G11+N.T11+N.A12"
## Warning: glm.fit: algorithm did not converge
## Warning: glm.fit: fitted rates numerically 0 occurred
## Warning: glm.fit: algorithm did not converge
## Warning: glm.fit: fitted rates numerically 0 occurred
## Warning in glm(regressionFormula, data = data, family = poisson(link
## = "log")): fitting to calculate the null deviance did not converge --
## increase 'maxit'?
## No shape parameters included in fit.
## An object of class 'model'
##
## Slot "name": AR-DBD R8 Nucleotides + Fixed Shape Values (Rev. Comp. Sym.)
## Slot "varRegLen": 23
## Slot "leftFixedSeq": GTTCAGAGTTCTACAGTCCGACGATC
## Slot "rightFixedSeq": TGGAATTCTCGGGTGCCAAGG
## Slot "leftFixedSeqOverlap": 5
## Slot "rightFixedSeqOverlap": 5
## Slot "confidenceLevel": 0.99
## Slot "minAffinity": 0.01
## Slot "missingValueSuppression": 0.5
## Slot "minSeedValue": 0.01
## Slot "seedLen": 15
## Slot "consensusSeq": [AG]G[AT]ACA[ACGT][ACGT][ACGT]TGT[AT]C[CT]
## Slot "upFootprintExtend": 4
## Slot "downFootprintExtend": 4
## Slot "fpLen": 23
##
## Fits a model of footprint length 23 for mono-nucleotide and shape features (shape = HelT, MGW) with fixed values for shape parameter positions not included in Shape.set used as offsets for the glm fit, and 11 view(s) per strand of DNA and 1 round(s) of data (round = 8) with reverse complement symmetry.
##
## Slot "regressionFormula": ObservedCount ~ offset(logProb)+offset(fixedSddG)+Round.8+N.A1+N.C1+N.G1+N.T1+N.A2+N.C2+N.G2+N.T2+N.A3+N.C3+N.G3+N.T3+N.A4+N.C4+N.G4+N.T4+N.A5+N.C5+N.G5+N.T5+N.A6+N.C6+N.G6+N.T6+N.A7+N.C7+N.G7+N.T7+N.A8+N.C8+N.G8+N.T8+N.A9+N.C9+N.G9+N.T9+N.A10+N.C10+N.G10+N.T10+N.A11+N.C11+N.G11+N.T11+N.A12+N.C12
##
## Slot "shapeParamsUsed[[1]]": HelT MGW
##
## Includes the following feature sub-classes:
## An object of class 'N'
## Fits 23 nucleotides for a feature model of length 23.
## Nucleotide features are reverse complement symmetric.
## Nucleotide beta values:
## 1 2 3 4 5
## N.A 0.006103358 0.00000000 0.00000000 0.00000000 0.00000000
## N.C 0.000000000 -0.05289212 -0.05511278 -0.14463839 -0.42369065
## N.G 0.007256250 0.02391278 -0.14326843 -0.06237793 -0.09609331
## N.T -0.036280412 -0.03447726 -0.03931120 -0.11660322 -0.60782364
## 6 7 8 9 10 11
## N.A -0.7752476 -0.8071446 0.0000000 -1.9506264 0.0000000 -0.34882272
## N.C -0.6285128 -1.1118885 -0.4373724 0.0000000 -0.9003735 0.00000000
## N.G 0.0000000 -0.2800461 -1.7099007 0.4529522 -0.5121286 -0.28594046
## N.T -0.7975488 0.0000000 -1.1882996 -0.5276423 -0.6138186 -0.08420791
## 12 13 14 15 16 17
## N.A -0.1058195 -0.08420791 -0.6138186 -0.5276423 -1.1882996 0.0000000
## N.C 0.0000000 -0.28594046 -0.5121286 0.4529522 -1.7099007 -0.2800461
## N.G 0.0000000 0.00000000 -0.9003735 0.0000000 -0.4373724 -1.1118885
## N.T -0.1058195 -0.34882272 0.0000000 -1.9506264 0.0000000 -0.8071446
## 18 19 20 21 22
## N.A -0.7975488 -0.60782364 -0.11660322 -0.03931120 -0.03447726
## N.C 0.0000000 -0.09609331 -0.06237793 -0.14326843 0.02391278
## N.G -0.6285128 -0.42369065 -0.14463839 -0.05511278 -0.05289212
## N.T -0.7752476 0.00000000 0.00000000 0.00000000 0.00000000
## 23
## N.A -0.036280412
## N.C 0.007256250
## N.G 0.000000000
## N.T 0.006103358
##
## Nucleotide beta errors:
## 1 2 3 4 5
## N.A 0.0001377786 0.0000000000 0.0000000000 0.0000000000 0.000000000
## N.C 0.0000000000 0.0001559732 0.0001771755 0.0002581687 0.001060001
## N.G 0.0001705080 0.0001452433 0.0002368831 0.0001626176 0.000147751
## N.T 0.0001706143 0.0001754550 0.0001695992 0.0001653681 0.000634667
## 6 7 8 9 10
## N.A 0.012317098 0.0001294108 0.00000000 0.02727699 0.0000000000
## N.C 0.019632769 0.0002962170 0.04166681 0.00000000 0.0228219778
## N.G 0.000000000 0.0003996975 0.01804277 0.05872959 0.0003192132
## N.T 0.001183016 0.0000000000 0.05734733 0.05896035 0.0049448411
## 11 12 13 14 15
## N.A 0.0004204734 0.0001783287 0.0001298898 0.0049448411 0.05896035
## N.C 0.0000000000 0.0000000000 0.0003274428 0.0003192132 0.05872959
## N.G 0.0003274428 0.0000000000 0.0000000000 0.0228219778 0.00000000
## N.T 0.0001298898 0.0001783287 0.0004204734 0.0000000000 0.02727699
## 16 17 18 19 20
## N.A 0.05734733 0.0000000000 0.001183016 0.000634667 0.0001653681
## N.C 0.01804277 0.0003996975 0.000000000 0.000147751 0.0001626176
## N.G 0.04166681 0.0002962170 0.019632769 0.001060001 0.0002581687
## N.T 0.00000000 0.0001294108 0.012317098 0.000000000 0.0000000000
## 21 22 23
## N.A 0.0001695992 0.0001754550 0.0001706143
## N.C 0.0002368831 0.0001452433 0.0001705080
## N.G 0.0001771755 0.0001559732 0.0000000000
## N.T 0.0000000000 0.0000000000 0.0001377786
##
##
## An object of class 'Intercept'
## Fits intercept(s) for 1 round(s) (round = 8).
## Intercept beta values:
## Round.8:
## [1] 39.36242
##
## Intercept beta errors:
## Round.8:
## [1] 0.00261764
##
##
##
## An object of class 'Shape'
## Fits 3 shape coefficients for 3 kinds of shape parameter(s) (shape = HelT, MGW) for a feature model of length 23.
## Shape features are reverse complement symmetric.
## Shape beta values:
## 1 2 3 4 5
## Shape.HelTA 0.012777702 0.009207855 0.01059514 0.02979068 -0.050599877
## Shape.HelTB 0.005339387 -0.003648404 -0.03570894 -0.01147808 -0.008140565
## Shape.MGW 0.011501160 -0.008920502 -0.02962055 -0.18939069 -0.087382882
## 6 7 8 9 10
## Shape.HelTA -0.04637238 -0.1065263 0.1929080 0.19583385 0.11049638
## Shape.HelTB -0.12908078 0.3710017 0.7437455 -0.03934332 -0.07502132
## Shape.MGW 0.06924589 0.3766392 1.2499150 0.17179767 -0.17550875
## 11 12 13 14 15
## Shape.HelTA 0.09218623 0.02171862 -0.03883268 -0.07502132 -0.03934332
## Shape.HelTB -0.03883268 0.02171862 0.09218623 0.11049638 0.19583385
## Shape.MGW -0.12381238 0.15027160 -0.12381238 -0.17550875 0.17179767
## 16 17 18 19 20
## Shape.HelTA 0.7437455 0.3710017 -0.12908078 -0.008140565 -0.01147808
## Shape.HelTB 0.1929080 -0.1065263 -0.04637238 -0.050599877 0.02979068
## Shape.MGW 1.2499150 0.3766392 0.06924589 -0.087382882 -0.18939069
## 21 22 23
## Shape.HelTA -0.03570894 -0.003648404 0.005339387
## Shape.HelTB 0.01059514 0.009207855 0.012777702
## Shape.MGW -0.02962055 -0.008920502 0.011501160
##
## Shape beta errors:
## 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23
## Shape.HelTA 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0
## Shape.HelTB 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0
## Shape.MGW 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0
##
## [1] "Number of Observations in Design Matrix: 992334"
## No shape parameters included in fit.
## [1] "i = 3"
## [1] "Round summary: "
## 8 Total
## Round 992334 992334
## [1] "Mono-nucleotide summary: "
## N.A N.C N.G N.T
## 1 578798 634475 345980 425415
## 2 713000 468119 369403 434146
## 3 904808 495879 168310 415671
## 4 922498 252340 405276 404554
## 5 1283737 17453 642160 41318
## 6 709 451 1968988 14520
## 7 835414 104455 93013 951786
## 8 1982214 674 947 833
## 9 588 1982631 858 591
## 10 1867153 706 113729 3080
## 11 79357 1105674 111273 688364
## 12 342383 649951 0 0
## [1] "View/strand orientation summary: "
## View.1 View.2 View.3 View.4 View.5 View.6 View.7 View.8 View.9
## Strand.F 5072 63812 96434 69453 106766 128861 140302 152573 110474
## Strand.R 0 0 0 0 0 0 0 0 0
## View.10 View.11 StrandTotal
## Strand.F 56629 61958 992334
## Strand.R 0 0 0
## [1] "Regression Formula: "
## [1] "ObservedCount ~ offset(logProb)+offset(fixedSddG)+N.A1+N.G1+N.T1+N.C2+N.G2+N.T2+N.C3+N.G3+N.T3+N.C4+N.G4+N.T4+N.C5+N.G5+N.T5+N.A6+N.C6+N.T6+N.A7+N.C7+N.G7+N.C8+N.G8+N.T8+N.A9+N.G9+N.T9+N.C10+N.G10+N.T10+N.A11+N.G11+N.T11+N.A12"
## Warning: glm.fit: fitted rates numerically 0 occurred
## No shape parameters included in fit.
## An object of class 'model'
##
## Slot "name": AR-DBD R8 Nucleotides + Fixed Shape Values (Rev. Comp. Sym.)
## Slot "varRegLen": 23
## Slot "leftFixedSeq": GTTCAGAGTTCTACAGTCCGACGATC
## Slot "rightFixedSeq": TGGAATTCTCGGGTGCCAAGG
## Slot "leftFixedSeqOverlap": 5
## Slot "rightFixedSeqOverlap": 5
## Slot "confidenceLevel": 0.99
## Slot "minAffinity": 0.01
## Slot "missingValueSuppression": 0.5
## Slot "minSeedValue": 0.01
## Slot "seedLen": 15
## Slot "consensusSeq": [AG]G[AT]ACA[ACGT][ACGT][ACGT]TGT[AT]C[CT]
## Slot "upFootprintExtend": 4
## Slot "downFootprintExtend": 4
## Slot "fpLen": 23
##
## Fits a model of footprint length 23 for mono-nucleotide and shape features (shape = HelT, MGW) with fixed values for shape parameter positions not included in Shape.set used as offsets for the glm fit, and 11 view(s) per strand of DNA and 1 round(s) of data (round = 8) with reverse complement symmetry.
##
## Slot "regressionFormula": ObservedCount ~ offset(logProb)+offset(fixedSddG)+Round.8+N.A1+N.C1+N.G1+N.T1+N.A2+N.C2+N.G2+N.T2+N.A3+N.C3+N.G3+N.T3+N.A4+N.C4+N.G4+N.T4+N.A5+N.C5+N.G5+N.T5+N.A6+N.C6+N.G6+N.T6+N.A7+N.C7+N.G7+N.T7+N.A8+N.C8+N.G8+N.T8+N.A9+N.C9+N.G9+N.T9+N.A10+N.C10+N.G10+N.T10+N.A11+N.C11+N.G11+N.T11+N.A12+N.C12
##
## Slot "shapeParamsUsed[[1]]": HelT MGW
##
## Includes the following feature sub-classes:
## An object of class 'N'
## Fits 23 nucleotides for a feature model of length 23.
## Nucleotide features are reverse complement symmetric.
## Nucleotide beta values:
## 1 2 3 4 5
## N.A 0.006078596 0.00000000 0.00000000 0.00000000 0.00000000
## N.C 0.000000000 -0.05285798 -0.05538693 -0.14499125 -0.42680932
## N.G 0.007207626 0.02394681 -0.14374003 -0.06255949 -0.09655342
## N.T -0.036265591 -0.03435153 -0.03940771 -0.11706989 -0.60888093
## 6 7 8 9 10 11
## N.A -0.7992291 -0.8075073 0.0000000 -1.9483814 0.0000000 -0.34890706
## N.C -0.6109298 -1.1121540 -0.5693159 0.0000000 -0.9715652 0.00000000
## N.G 0.0000000 -0.2803594 -1.7261507 0.1461386 -0.5121385 -0.28600896
## N.T -0.8013201 0.0000000 -0.9583589 -0.6986228 -0.6137702 -0.08438165
## 12 13 14 15 16 17
## N.A -0.1056784 -0.08438165 -0.6137702 -0.6986228 -0.9583589 0.0000000
## N.C 0.0000000 -0.28600896 -0.5121385 0.1461386 -1.7261507 -0.2803594
## N.G 0.0000000 0.00000000 -0.9715652 0.0000000 -0.5693159 -1.1121540
## N.T -0.1056784 -0.34890706 0.0000000 -1.9483814 0.0000000 -0.8075073
## 18 19 20 21 22
## N.A -0.8013201 -0.60888093 -0.11706989 -0.03940771 -0.03435153
## N.C 0.0000000 -0.09655342 -0.06255949 -0.14374003 0.02394681
## N.G -0.6109298 -0.42680932 -0.14499125 -0.05538693 -0.05285798
## N.T -0.7992291 0.00000000 0.00000000 0.00000000 0.00000000
## 23
## N.A -0.036265591
## N.C 0.007207626
## N.G 0.000000000
## N.T 0.006078596
##
## Nucleotide beta errors:
## 1 2 3 4 5
## N.A 0.0001378775 0.0000000000 0.0000000000 0.0000000000 0.0000000000
## N.C 0.0000000000 0.0001560677 0.0001773752 0.0002584840 0.0010816572
## N.G 0.0001706382 0.0001453533 0.0002374205 0.0001627232 0.0001478749
## N.T 0.0001707187 0.0001755608 0.0001697169 0.0001655874 0.0006376307
## 6 7 8 9 10
## N.A 0.013558950 0.0001295034 0.00000000 0.02727747 0.0000000000
## N.C 0.016753870 0.0002963470 0.05125807 0.00000000 0.0303171816
## N.G 0.000000000 0.0003998916 0.01928846 0.05018130 0.0003195155
## N.T 0.001199812 0.0000000000 0.05093628 0.05948299 0.0049351917
## 11 12 13 14 15
## N.A 0.0004206859 0.000178391 0.0001299858 0.0049351917 0.05948299
## N.C 0.0000000000 0.000000000 0.0003275073 0.0003195155 0.05018130
## N.G 0.0003275073 0.000000000 0.0000000000 0.0303171816 0.00000000
## N.T 0.0001299858 0.000178391 0.0004206859 0.0000000000 0.02727747
## 16 17 18 19 20
## N.A 0.05093628 0.0000000000 0.001199812 0.0006376307 0.0001655874
## N.C 0.01928846 0.0003998916 0.000000000 0.0001478749 0.0001627232
## N.G 0.05125807 0.0002963470 0.016753870 0.0010816572 0.0002584840
## N.T 0.00000000 0.0001295034 0.013558950 0.0000000000 0.0000000000
## 21 22 23
## N.A 0.0001697169 0.0001755608 0.0001707187
## N.C 0.0002374205 0.0001453533 0.0001706382
## N.G 0.0001773752 0.0001560677 0.0000000000
## N.T 0.0000000000 0.0000000000 0.0001378775
##
##
## An object of class 'Intercept'
## Fits intercept(s) for 1 round(s) (round = 8).
## Intercept beta values:
## Round.8:
## [1] 39.36999
##
## Intercept beta errors:
## Round.8:
## [1] 0.002618464
##
##
##
## An object of class 'Shape'
## Fits 3 shape coefficients for 3 kinds of shape parameter(s) (shape = HelT, MGW) for a feature model of length 23.
## Shape features are reverse complement symmetric.
## Shape beta values:
## 1 2 3 4 5
## Shape.HelTA 0.012777702 0.009207855 0.01059514 0.02979068 -0.050599877
## Shape.HelTB 0.005339387 -0.003648404 -0.03570894 -0.01147808 -0.008140565
## Shape.MGW 0.011501160 -0.008920502 -0.02962055 -0.18939069 -0.087382882
## 6 7 8 9 10
## Shape.HelTA -0.04637238 -0.1065263 0.1929080 0.19583385 0.11049638
## Shape.HelTB -0.12908078 0.3710017 0.7437455 -0.03934332 -0.07502132
## Shape.MGW 0.06924589 0.3766392 1.2499150 0.17179767 -0.17550875
## 11 12 13 14 15
## Shape.HelTA 0.09218623 0.02171862 -0.03883268 -0.07502132 -0.03934332
## Shape.HelTB -0.03883268 0.02171862 0.09218623 0.11049638 0.19583385
## Shape.MGW -0.12381238 0.15027160 -0.12381238 -0.17550875 0.17179767
## 16 17 18 19 20
## Shape.HelTA 0.7437455 0.3710017 -0.12908078 -0.008140565 -0.01147808
## Shape.HelTB 0.1929080 -0.1065263 -0.04637238 -0.050599877 0.02979068
## Shape.MGW 1.2499150 0.3766392 0.06924589 -0.087382882 -0.18939069
## 21 22 23
## Shape.HelTA -0.03570894 -0.003648404 0.005339387
## Shape.HelTB 0.01059514 0.009207855 0.012777702
## Shape.MGW -0.02962055 -0.008920502 0.011501160
##
## Shape beta errors:
## 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23
## Shape.HelTA 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0
## Shape.HelTB 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0
## Shape.MGW 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0
##
## [1] "Number of Observations in Design Matrix: 992117"
## No shape parameters included in fit.
## [1] "i = 4"
## [1] "Round summary: "
## 8 Total
## Round 992117 992117
## [1] "Mono-nucleotide summary: "
## N.A N.C N.G N.T
## 1 578655 634393 345895 425291
## 2 712816 468041 369304 434073
## 3 904715 495755 168157 415607
## 4 922297 252245 405233 404459
## 5 1283470 17394 642100 41270
## 6 668 275 1968811 14480
## 7 835228 104451 92941 951614
## 8 1982155 358 943 778
## 9 583 1982621 450 580
## 10 1866901 697 113700 2936
## 11 79183 1105661 111230 688160
## 12 342350 649767 0 0
## [1] "View/strand orientation summary: "
## View.1 View.2 View.3 View.4 View.5 View.6 View.7 View.8 View.9
## Strand.F 5062 63784 96422 69438 106742 128834 140284 152557 110458
## Strand.R 0 0 0 0 0 0 0 0 0
## View.10 View.11 StrandTotal
## Strand.F 56608 61928 992117
## Strand.R 0 0 0
## [1] "Regression Formula: "
## [1] "ObservedCount ~ offset(logProb)+offset(fixedSddG)+N.A1+N.G1+N.T1+N.C2+N.G2+N.T2+N.C3+N.G3+N.T3+N.C4+N.G4+N.T4+N.C5+N.G5+N.T5+N.A6+N.C6+N.T6+N.A7+N.C7+N.G7+N.C8+N.G8+N.T8+N.A9+N.G9+N.T9+N.C10+N.G10+N.T10+N.A11+N.G11+N.T11+N.A12"
## Warning: glm.fit: fitted rates numerically 0 occurred
## Warning: glm.fit: fitted rates numerically 0 occurred
## No shape parameters included in fit.
## An object of class 'model'
##
## Slot "name": AR-DBD R8 Nucleotides + Fixed Shape Values (Rev. Comp. Sym.)
## Slot "varRegLen": 23
## Slot "leftFixedSeq": GTTCAGAGTTCTACAGTCCGACGATC
## Slot "rightFixedSeq": TGGAATTCTCGGGTGCCAAGG
## Slot "leftFixedSeqOverlap": 5
## Slot "rightFixedSeqOverlap": 5
## Slot "confidenceLevel": 0.99
## Slot "minAffinity": 0.01
## Slot "missingValueSuppression": 0.5
## Slot "minSeedValue": 0.01
## Slot "seedLen": 15
## Slot "consensusSeq": [AG]G[AT]ACA[ACGT][ACGT][ACGT]TGT[AT]C[CT]
## Slot "upFootprintExtend": 4
## Slot "downFootprintExtend": 4
## Slot "fpLen": 23
##
## Fits a model of footprint length 23 for mono-nucleotide and shape features (shape = HelT, MGW) with fixed values for shape parameter positions not included in Shape.set used as offsets for the glm fit, and 11 view(s) per strand of DNA and 1 round(s) of data (round = 8) with reverse complement symmetry.
##
## Slot "regressionFormula": ObservedCount ~ offset(logProb)+offset(fixedSddG)+Round.8+N.A1+N.C1+N.G1+N.T1+N.A2+N.C2+N.G2+N.T2+N.A3+N.C3+N.G3+N.T3+N.A4+N.C4+N.G4+N.T4+N.A5+N.C5+N.G5+N.T5+N.A6+N.C6+N.G6+N.T6+N.A7+N.C7+N.G7+N.T7+N.A8+N.C8+N.G8+N.T8+N.A9+N.C9+N.G9+N.T9+N.A10+N.C10+N.G10+N.T10+N.A11+N.C11+N.G11+N.T11+N.A12+N.C12
##
## Slot "shapeParamsUsed[[1]]": HelT MGW
##
## Includes the following feature sub-classes:
## An object of class 'N'
## Fits 23 nucleotides for a feature model of length 23.
## Nucleotide features are reverse complement symmetric.
## Nucleotide beta values:
## 1 2 3 4 5 6
## N.A 0.006077568 0.00000000 0.00000000 0.00000000 0.0000000 -0.7995496
## N.C 0.000000000 -0.05285850 -0.05537880 -0.14499029 -0.4265195 -0.6975009
## N.G 0.007208404 0.02394527 -0.14374535 -0.06255652 -0.0965528 0.0000000
## N.T -0.036262319 -0.03435356 -0.03940367 -0.11707599 -0.6088717 -0.8011972
## 7 8 9 10 11 12
## N.A -0.8075087 0.000000 -1.9483894 0.0000000 -0.34889908 -0.1056801
## N.C -1.1121536 -1.102018 0.0000000 -0.9714363 0.00000000 0.0000000
## N.G -0.2803507 -1.726147 0.6530464 -0.5121407 -0.28600519 0.0000000
## N.T 0.0000000 -0.739690 -1.5823726 -0.6128446 -0.08437955 -0.1056801
## 13 14 15 16 17 18
## N.A -0.08437955 -0.6128446 -1.5823726 -0.739690 0.0000000 -0.8011972
## N.C -0.28600519 -0.5121407 0.6530464 -1.726147 -0.2803507 0.0000000
## N.G 0.00000000 -0.9714363 0.0000000 -1.102018 -1.1121536 -0.6975009
## N.T -0.34889908 0.0000000 -1.9483894 0.000000 -0.8075087 -0.7995496
## 19 20 21 22 23
## N.A -0.6088717 -0.11707599 -0.03940367 -0.03435356 -0.036262319
## N.C -0.0965528 -0.06255652 -0.14374535 0.02394527 0.007208404
## N.G -0.4265195 -0.14499029 -0.05537880 -0.05285850 0.000000000
## N.T 0.0000000 0.00000000 0.00000000 0.00000000 0.006077568
##
## Nucleotide beta errors:
## 1 2 3 4 5
## N.A 0.0001378782 0.0000000000 0.0000000000 0.0000000000 0.0000000000
## N.C 0.0000000000 0.0001560686 0.0001773760 0.0002584855 0.0010811413
## N.G 0.0001706386 0.0001453544 0.0002374291 0.0001627242 0.0001478759
## N.T 0.0001707187 0.0001755629 0.0001697171 0.0001655913 0.0006376543
## 6 7 8 9 10
## N.A 0.013559407 0.0001295043 0.00000000 0.02727757 0.000000000
## N.C 0.026001922 0.0002963472 0.04008662 0.00000000 0.030317178
## N.G 0.000000000 0.0003999249 0.01928846 0.03233685 0.000319518
## N.T 0.001199819 0.0000000000 0.02903231 0.06367150 0.004947338
## 11 12 13 14 15
## N.A 0.0004206813 0.0001783911 0.0001299862 0.004947338 0.06367150
## N.C 0.0000000000 0.0000000000 0.0003275132 0.000319518 0.03233685
## N.G 0.0003275132 0.0000000000 0.0000000000 0.030317178 0.00000000
## N.T 0.0001299862 0.0001783911 0.0004206813 0.000000000 0.02727757
## 16 17 18 19 20
## N.A 0.02903231 0.0000000000 0.001199819 0.0006376543 0.0001655913
## N.C 0.01928846 0.0003999249 0.000000000 0.0001478759 0.0001627242
## N.G 0.04008662 0.0002963472 0.026001922 0.0010811413 0.0002584855
## N.T 0.00000000 0.0001295043 0.013559407 0.0000000000 0.0000000000
## 21 22 23
## N.A 0.0001697171 0.0001755629 0.0001707187
## N.C 0.0002374291 0.0001453544 0.0001706386
## N.G 0.0001773760 0.0001560686 0.0000000000
## N.T 0.0000000000 0.0000000000 0.0001378782
##
##
## An object of class 'Intercept'
## Fits intercept(s) for 1 round(s) (round = 8).
## Intercept beta values:
## Round.8:
## [1] 39.36996
##
## Intercept beta errors:
## Round.8:
## [1] 0.002618458
##
##
##
## An object of class 'Shape'
## Fits 3 shape coefficients for 3 kinds of shape parameter(s) (shape = HelT, MGW) for a feature model of length 23.
## Shape features are reverse complement symmetric.
## Shape beta values:
## 1 2 3 4 5
## Shape.HelTA 0.012777702 0.009207855 0.01059514 0.02979068 -0.050599877
## Shape.HelTB 0.005339387 -0.003648404 -0.03570894 -0.01147808 -0.008140565
## Shape.MGW 0.011501160 -0.008920502 -0.02962055 -0.18939069 -0.087382882
## 6 7 8 9 10
## Shape.HelTA -0.04637238 -0.1065263 0.1929080 0.19583385 0.11049638
## Shape.HelTB -0.12908078 0.3710017 0.7437455 -0.03934332 -0.07502132
## Shape.MGW 0.06924589 0.3766392 1.2499150 0.17179767 -0.17550875
## 11 12 13 14 15
## Shape.HelTA 0.09218623 0.02171862 -0.03883268 -0.07502132 -0.03934332
## Shape.HelTB -0.03883268 0.02171862 0.09218623 0.11049638 0.19583385
## Shape.MGW -0.12381238 0.15027160 -0.12381238 -0.17550875 0.17179767
## 16 17 18 19 20
## Shape.HelTA 0.7437455 0.3710017 -0.12908078 -0.008140565 -0.01147808
## Shape.HelTB 0.1929080 -0.1065263 -0.04637238 -0.050599877 0.02979068
## Shape.MGW 1.2499150 0.3766392 0.06924589 -0.087382882 -0.18939069
## 21 22 23
## Shape.HelTA -0.03570894 -0.003648404 0.005339387
## Shape.HelTB 0.01059514 0.009207855 0.012777702
## Shape.MGW -0.02962055 -0.008920502 0.011501160
##
## Shape beta errors:
## 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23
## Shape.HelTA 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0
## Shape.HelTB 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0
## Shape.MGW 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0
##
## [1] "Number of Observations in Design Matrix: 988927"
## No shape parameters included in fit.
## [1] "i = 5"
## [1] "Round summary: "
## 8 Total
## Round 988927 988927
## [1] "Mono-nucleotide summary: "
## N.A N.C N.G N.T
## 1 576765 632850 344675 423564
## 2 710277 466852 368516 432209
## 3 902683 494133 166656 414382
## 4 919627 251279 404224 402724
## 5 1280324 17325 639272 40933
## 6 654 255 1962601 14344
## 7 831518 104181 92629 949526
## 8 1975861 321 920 752
## 9 574 1976393 435 452
## 10 1860824 685 113443 2902
## 11 78268 1102207 110981 686398
## 12 341378 647549 0 0
## [1] "View/strand orientation summary: "
## View.1 View.2 View.3 View.4 View.5 View.6 View.7 View.8 View.9
## Strand.F 5043 63425 96053 69219 106303 128672 139833 152054 110214
## Strand.R 0 0 0 0 0 0 0 0 0
## View.10 View.11 StrandTotal
## Strand.F 56418 61693 988927
## Strand.R 0 0 0
## [1] "Regression Formula: "
## [1] "ObservedCount ~ offset(logProb)+offset(fixedSddG)+N.A1+N.G1+N.T1+N.C2+N.G2+N.T2+N.C3+N.G3+N.T3+N.C4+N.G4+N.T4+N.C5+N.G5+N.T5+N.A6+N.C6+N.T6+N.A7+N.C7+N.G7+N.C8+N.G8+N.T8+N.A9+N.G9+N.T9+N.C10+N.G10+N.T10+N.A11+N.G11+N.T11+N.A12"
## Warning: glm.fit: fitted rates numerically 0 occurred
## No shape parameters included in fit.
## An object of class 'model'
##
## Slot "name": AR-DBD R8 Nucleotides + Fixed Shape Values (Rev. Comp. Sym.)
## Slot "varRegLen": 23
## Slot "leftFixedSeq": GTTCAGAGTTCTACAGTCCGACGATC
## Slot "rightFixedSeq": TGGAATTCTCGGGTGCCAAGG
## Slot "leftFixedSeqOverlap": 5
## Slot "rightFixedSeqOverlap": 5
## Slot "confidenceLevel": 0.99
## Slot "minAffinity": 0.01
## Slot "missingValueSuppression": 0.5
## Slot "minSeedValue": 0.01
## Slot "seedLen": 15
## Slot "consensusSeq": [AG]G[AT]ACA[ACGT][ACGT][ACGT]TGT[AT]C[CT]
## Slot "upFootprintExtend": 4
## Slot "downFootprintExtend": 4
## Slot "fpLen": 23
##
## Fits a model of footprint length 23 for mono-nucleotide and shape features (shape = HelT, MGW) with fixed values for shape parameter positions not included in Shape.set used as offsets for the glm fit, and 11 view(s) per strand of DNA and 1 round(s) of data (round = 8) with reverse complement symmetry.
##
## Slot "regressionFormula": ObservedCount ~ offset(logProb)+offset(fixedSddG)+Round.8+N.A1+N.C1+N.G1+N.T1+N.A2+N.C2+N.G2+N.T2+N.A3+N.C3+N.G3+N.T3+N.A4+N.C4+N.G4+N.T4+N.A5+N.C5+N.G5+N.T5+N.A6+N.C6+N.G6+N.T6+N.A7+N.C7+N.G7+N.T7+N.A8+N.C8+N.G8+N.T8+N.A9+N.C9+N.G9+N.T9+N.A10+N.C10+N.G10+N.T10+N.A11+N.C11+N.G11+N.T11+N.A12+N.C12
##
## Slot "shapeParamsUsed[[1]]": HelT MGW
##
## Includes the following feature sub-classes:
## An object of class 'N'
## Fits 23 nucleotides for a feature model of length 23.
## Nucleotide features are reverse complement symmetric.
## Nucleotide beta values:
## 1 2 3 4 5 6
## N.A 0.006065039 0.00000000 0.00000000 0.0000000 0.00000000 -0.7989972
## N.C 0.000000000 -0.05282577 -0.05542611 -0.1450768 -0.42644547 -0.7779214
## N.G 0.007209796 0.02397266 -0.14384247 -0.0626175 -0.09673257 0.0000000
## N.T -0.036278854 -0.03431445 -0.03943247 -0.1172086 -0.60927650 -0.8017470
## 7 8 9 10 11 12
## N.A -0.8076300 0.0000000 -2.0085642 0.0000000 -0.34950410 -0.1056982
## N.C -1.1122210 -1.2182972 0.0000000 -1.0509511 0.00000000 0.0000000
## N.G -0.2804762 -1.7420307 0.7435719 -0.5121817 -0.28599711 0.0000000
## N.T 0.0000000 -0.6770943 -1.9540027 -0.6155590 -0.08439218 -0.1056982
## 13 14 15 16 17 18
## N.A -0.08439218 -0.6155590 -1.9540027 -0.6770943 0.0000000 -0.8017470
## N.C -0.28599711 -0.5121817 0.7435719 -1.7420307 -0.2804762 0.0000000
## N.G 0.00000000 -1.0509511 0.0000000 -1.2182972 -1.1122210 -0.7779214
## N.T -0.34950410 0.0000000 -2.0085642 0.0000000 -0.8076300 -0.7989972
## 19 20 21 22 23
## N.A -0.60927650 -0.1172086 -0.03943247 -0.03431445 -0.036278854
## N.C -0.09673257 -0.0626175 -0.14384247 0.02397266 0.007209796
## N.G -0.42644547 -0.1450768 -0.05542611 -0.05282577 0.000000000
## N.T 0.00000000 0.0000000 0.00000000 0.00000000 0.006065039
##
## Nucleotide beta errors:
## 1 2 3 4 5
## N.A 0.0001379289 0.0000000000 0.0000000000 0.0000000000 0.0000000000
## N.C 0.0000000000 0.0001561209 0.0001774569 0.0002586165 0.0010809722
## N.G 0.0001706915 0.0001454015 0.0002377685 0.0001627889 0.0001479497
## N.T 0.0001707965 0.0001756681 0.0001697708 0.0001656763 0.0006390689
## 6 7 8 9 10
## N.A 0.013481926 0.0001295572 0.00000000 0.03466907 0.0000000000
## N.C 0.036084558 0.0002964438 0.03656940 0.00000000 0.0416668326
## N.G 0.000000000 0.0004001387 0.02055038 0.02149963 0.0003196237
## N.T 0.001202279 0.0000000000 0.02101136 0.06265854 0.0050005480
## 11 12 13 14 15
## N.A 0.0004225257 0.000178446 0.0001300260 0.0050005480 0.06265854
## N.C 0.0000000000 0.000000000 0.0003276029 0.0003196237 0.02149963
## N.G 0.0003276029 0.000000000 0.0000000000 0.0416668326 0.00000000
## N.T 0.0001300260 0.000178446 0.0004225257 0.0000000000 0.03466907
## 16 17 18 19 20
## N.A 0.02101136 0.0000000000 0.001202279 0.0006390689 0.0001656763
## N.C 0.02055038 0.0004001387 0.000000000 0.0001479497 0.0001627889
## N.G 0.03656940 0.0002964438 0.036084558 0.0010809722 0.0002586165
## N.T 0.00000000 0.0001295572 0.013481926 0.0000000000 0.0000000000
## 21 22 23
## N.A 0.0001697708 0.0001756681 0.0001707965
## N.C 0.0002377685 0.0001454015 0.0001706915
## N.G 0.0001774569 0.0001561209 0.0000000000
## N.T 0.0000000000 0.0000000000 0.0001379289
##
##
## An object of class 'Intercept'
## Fits intercept(s) for 1 round(s) (round = 8).
## Intercept beta values:
## Round.8:
## [1] 39.37224
##
## Intercept beta errors:
## Round.8:
## [1] 0.0026191
##
##
##
## An object of class 'Shape'
## Fits 3 shape coefficients for 3 kinds of shape parameter(s) (shape = HelT, MGW) for a feature model of length 23.
## Shape features are reverse complement symmetric.
## Shape beta values:
## 1 2 3 4 5
## Shape.HelTA 0.012777702 0.009207855 0.01059514 0.02979068 -0.050599877
## Shape.HelTB 0.005339387 -0.003648404 -0.03570894 -0.01147808 -0.008140565
## Shape.MGW 0.011501160 -0.008920502 -0.02962055 -0.18939069 -0.087382882
## 6 7 8 9 10
## Shape.HelTA -0.04637238 -0.1065263 0.1929080 0.19583385 0.11049638
## Shape.HelTB -0.12908078 0.3710017 0.7437455 -0.03934332 -0.07502132
## Shape.MGW 0.06924589 0.3766392 1.2499150 0.17179767 -0.17550875
## 11 12 13 14 15
## Shape.HelTA 0.09218623 0.02171862 -0.03883268 -0.07502132 -0.03934332
## Shape.HelTB -0.03883268 0.02171862 0.09218623 0.11049638 0.19583385
## Shape.MGW -0.12381238 0.15027160 -0.12381238 -0.17550875 0.17179767
## 16 17 18 19 20
## Shape.HelTA 0.7437455 0.3710017 -0.12908078 -0.008140565 -0.01147808
## Shape.HelTB 0.1929080 -0.1065263 -0.04637238 -0.050599877 0.02979068
## Shape.MGW 1.2499150 0.3766392 0.06924589 -0.087382882 -0.18939069
## 21 22 23
## Shape.HelTA -0.03570894 -0.003648404 0.005339387
## Shape.HelTB 0.01059514 0.009207855 0.012777702
## Shape.MGW -0.02962055 -0.008920502 0.011501160
##
## Shape beta errors:
## 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23
## Shape.HelTA 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0
## Shape.HelTB 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0
## Shape.MGW 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0
##
## [1] "Number of Observations in Design Matrix: 986714"
## No shape parameters included in fit.
## [1] "i = 6"
## [1] "Round summary: "
## 8 Total
## Round 986714 986714
## [1] "Mono-nucleotide summary: "
## N.A N.C N.G N.T
## 1 575376 631750 343781 422521
## 2 708711 466029 367824 430864
## 3 901319 493117 165635 413357
## 4 917936 250668 403565 401259
## 5 1278197 17288 637232 40711
## 6 645 249 1958246 14288
## 7 829029 104030 92384 947985
## 8 1971465 313 912 738
## 9 570 1972064 428 366
## 10 1856554 681 113305 2888
## 11 77929 1099748 110830 684921
## 12 340321 646393 0 0
## [1] "View/strand orientation summary: "
## View.1 View.2 View.3 View.4 View.5 View.6 View.7 View.8 View.9
## Strand.F 5031 63204 95790 69060 106024 128507 139449 151760 110038
## Strand.R 0 0 0 0 0 0 0 0 0
## View.10 View.11 StrandTotal
## Strand.F 56303 61548 986714
## Strand.R 0 0 0
## [1] "Regression Formula: "
## [1] "ObservedCount ~ offset(logProb)+offset(fixedSddG)+N.A1+N.G1+N.T1+N.C2+N.G2+N.T2+N.C3+N.G3+N.T3+N.C4+N.G4+N.T4+N.C5+N.G5+N.T5+N.A6+N.C6+N.T6+N.A7+N.C7+N.G7+N.C8+N.G8+N.T8+N.A9+N.G9+N.T9+N.C10+N.G10+N.T10+N.A11+N.G11+N.T11+N.A12"
## Warning: glm.fit: fitted rates numerically 0 occurred
## No shape parameters included in fit.
## An object of class 'model'
##
## Slot "name": AR-DBD R8 Nucleotides + Fixed Shape Values (Rev. Comp. Sym.)
## Slot "varRegLen": 23
## Slot "leftFixedSeq": GTTCAGAGTTCTACAGTCCGACGATC
## Slot "rightFixedSeq": TGGAATTCTCGGGTGCCAAGG
## Slot "leftFixedSeqOverlap": 5
## Slot "rightFixedSeqOverlap": 5
## Slot "confidenceLevel": 0.99
## Slot "minAffinity": 0.01
## Slot "missingValueSuppression": 0.5
## Slot "minSeedValue": 0.01
## Slot "seedLen": 15
## Slot "consensusSeq": [AG]G[AT]ACA[ACGT][ACGT][ACGT]TGT[AT]C[CT]
## Slot "upFootprintExtend": 4
## Slot "downFootprintExtend": 4
## Slot "fpLen": 23
##
## Fits a model of footprint length 23 for mono-nucleotide and shape features (shape = HelT, MGW) with fixed values for shape parameter positions not included in Shape.set used as offsets for the glm fit, and 11 view(s) per strand of DNA and 1 round(s) of data (round = 8) with reverse complement symmetry.
##
## Slot "regressionFormula": ObservedCount ~ offset(logProb)+offset(fixedSddG)+Round.8+N.A1+N.C1+N.G1+N.T1+N.A2+N.C2+N.G2+N.T2+N.A3+N.C3+N.G3+N.T3+N.A4+N.C4+N.G4+N.T4+N.A5+N.C5+N.G5+N.T5+N.A6+N.C6+N.G6+N.T6+N.A7+N.C7+N.G7+N.T7+N.A8+N.C8+N.G8+N.T8+N.A9+N.C9+N.G9+N.T9+N.A10+N.C10+N.G10+N.T10+N.A11+N.C11+N.G11+N.T11+N.A12+N.C12
##
## Slot "shapeParamsUsed[[1]]": HelT MGW
##
## Includes the following feature sub-classes:
## An object of class 'N'
## Fits 23 nucleotides for a feature model of length 23.
## Nucleotide features are reverse complement symmetric.
## Nucleotide beta values:
## 1 2 3 4 5
## N.A 0.006040944 0.00000000 0.00000000 0.00000000 0.00000000
## N.C 0.000000000 -0.05279784 -0.05543689 -0.14516382 -0.42643495
## N.G 0.007218393 0.02397867 -0.14383838 -0.06265629 -0.09687191
## N.T -0.036283332 -0.03427430 -0.03945393 -0.11732245 -0.60963045
## 6 7 8 9 10 11
## N.A -0.8004893 -0.8077304 0.0000000 -2.0071724 0.0000000 -0.34974760
## N.C -0.7779108 -1.1122544 -1.2187106 0.0000000 -1.0509815 0.00000000
## N.G 0.0000000 -0.2805716 -1.7420457 0.7441363 -0.5121789 -0.28600896
## N.T -0.8018132 0.0000000 -0.6716486 -1.9879498 -0.6155583 -0.08441039
## 12 13 14 15 16 17
## N.A -0.1057471 -0.08441039 -0.6155583 -1.9879498 -0.6716486 0.0000000
## N.C 0.0000000 -0.28600896 -0.5121789 0.7441363 -1.7420457 -0.2805716
## N.G 0.0000000 0.00000000 -1.0509815 0.0000000 -1.2187106 -1.1122544
## N.T -0.1057471 -0.34974760 0.0000000 -2.0071724 0.0000000 -0.8077304
## 18 19 20 21 22
## N.A -0.8018132 -0.60963045 -0.11732245 -0.03945393 -0.03427430
## N.C 0.0000000 -0.09687191 -0.06265629 -0.14383838 0.02397867
## N.G -0.7779108 -0.42643495 -0.14516382 -0.05543689 -0.05279784
## N.T -0.8004893 0.00000000 0.00000000 0.00000000 0.00000000
## 23
## N.A -0.036283332
## N.C 0.007218393
## N.G 0.000000000
## N.T 0.006040944
##
## Nucleotide beta errors:
## 1 2 3 4 5
## N.A 0.0001379754 0.0000000000 0.0000000000 0.0000000000 0.0000000000
## N.C 0.0000000000 0.0001561662 0.0001775189 0.0002587256 0.0010811715
## N.G 0.0001707335 0.0001454432 0.0002380312 0.0001628362 0.0001480137
## N.T 0.0001708624 0.0001757644 0.0001698287 0.0001657529 0.0006402737
## 6 7 8 9 10
## N.A 0.01356108 0.0001296018 0.00000000 0.03466908 0.0000000000
## N.C 0.03608456 0.0002964923 0.03645032 0.00000000 0.0416668328
## N.G 0.00000000 0.0004002757 0.02055038 0.02129569 0.0003196875
## N.T 0.00120320 0.0000000000 0.02060052 0.06596692 0.0050005604
## 11 12 13 14 15
## N.A 0.0004233749 0.0001785151 0.0001300609 0.0050005604 0.06596692
## N.C 0.0000000000 0.0000000000 0.0003276749 0.0003196875 0.02129569
## N.G 0.0003276749 0.0000000000 0.0000000000 0.0416668328 0.00000000
## N.T 0.0001300609 0.0001785151 0.0004233749 0.0000000000 0.03466908
## 16 17 18 19 20
## N.A 0.02060052 0.0000000000 0.00120320 0.0006402737 0.0001657529
## N.C 0.02055038 0.0004002757 0.00000000 0.0001480137 0.0001628362
## N.G 0.03645032 0.0002964923 0.03608456 0.0010811715 0.0002587256
## N.T 0.00000000 0.0001296018 0.01356108 0.0000000000 0.0000000000
## 21 22 23
## N.A 0.0001698287 0.0001757644 0.0001708624
## N.C 0.0002380312 0.0001454432 0.0001707335
## N.G 0.0001775189 0.0001561662 0.0000000000
## N.T 0.0000000000 0.0000000000 0.0001379754
##
##
## An object of class 'Intercept'
## Fits intercept(s) for 1 round(s) (round = 8).
## Intercept beta values:
## Round.8:
## [1] 39.374
##
## Intercept beta errors:
## Round.8:
## [1] 0.002619743
##
##
##
## An object of class 'Shape'
## Fits 3 shape coefficients for 3 kinds of shape parameter(s) (shape = HelT, MGW) for a feature model of length 23.
## Shape features are reverse complement symmetric.
## Shape beta values:
## 1 2 3 4 5
## Shape.HelTA 0.012777702 0.009207855 0.01059514 0.02979068 -0.050599877
## Shape.HelTB 0.005339387 -0.003648404 -0.03570894 -0.01147808 -0.008140565
## Shape.MGW 0.011501160 -0.008920502 -0.02962055 -0.18939069 -0.087382882
## 6 7 8 9 10
## Shape.HelTA -0.04637238 -0.1065263 0.1929080 0.19583385 0.11049638
## Shape.HelTB -0.12908078 0.3710017 0.7437455 -0.03934332 -0.07502132
## Shape.MGW 0.06924589 0.3766392 1.2499150 0.17179767 -0.17550875
## 11 12 13 14 15
## Shape.HelTA 0.09218623 0.02171862 -0.03883268 -0.07502132 -0.03934332
## Shape.HelTB -0.03883268 0.02171862 0.09218623 0.11049638 0.19583385
## Shape.MGW -0.12381238 0.15027160 -0.12381238 -0.17550875 0.17179767
## 16 17 18 19 20
## Shape.HelTA 0.7437455 0.3710017 -0.12908078 -0.008140565 -0.01147808
## Shape.HelTB 0.1929080 -0.1065263 -0.04637238 -0.050599877 0.02979068
## Shape.MGW 1.2499150 0.3766392 0.06924589 -0.087382882 -0.18939069
## 21 22 23
## Shape.HelTA -0.03570894 -0.003648404 0.005339387
## Shape.HelTB 0.01059514 0.009207855 0.012777702
## Shape.MGW -0.02962055 -0.008920502 0.011501160
##
## Shape beta errors:
## 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23
## Shape.HelTA 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0
## Shape.HelTB 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0
## Shape.MGW 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0
##
## [1] "Number of Observations in Design Matrix: 986604"
## No shape parameters included in fit.
## [1] "i = 7"
## [1] "Round summary: "
## 8 Total
## Round 986604 986604
## [1] "Mono-nucleotide summary: "
## N.A N.C N.G N.T
## 1 575309 631681 343742 422476
## 2 708615 465993 367793 430807
## 3 901239 493080 165580 413309
## 4 917842 250637 403535 401194
## 5 1278080 17287 637145 40696
## 6 645 249 1958029 14285
## 7 828916 104025 92376 947891
## 8 1971246 312 912 738
## 9 570 1971849 428 361
## 10 1856341 681 113298 2888
## 11 77914 1099620 110817 684857
## 12 340280 646324 0 0
## [1] "View/strand orientation summary: "
## View.1 View.2 View.3 View.4 View.5 View.6 View.7 View.8 View.9
## Strand.F 5030 63192 95778 69057 106009 128500 139431 151738 110027
## Strand.R 0 0 0 0 0 0 0 0 0
## View.10 View.11 StrandTotal
## Strand.F 56300 61542 986604
## Strand.R 0 0 0
## [1] "Regression Formula: "
## [1] "ObservedCount ~ offset(logProb)+offset(fixedSddG)+N.A1+N.G1+N.T1+N.C2+N.G2+N.T2+N.C3+N.G3+N.T3+N.C4+N.G4+N.T4+N.C5+N.G5+N.T5+N.A6+N.C6+N.T6+N.A7+N.C7+N.G7+N.C8+N.G8+N.T8+N.A9+N.G9+N.T9+N.C10+N.G10+N.T10+N.A11+N.G11+N.T11+N.A12"
## Warning: glm.fit: fitted rates numerically 0 occurred
## No shape parameters included in fit.
## An object of class 'model'
##
## Slot "name": AR-DBD R8 Nucleotides + Fixed Shape Values (Rev. Comp. Sym.)
## Slot "varRegLen": 23
## Slot "leftFixedSeq": GTTCAGAGTTCTACAGTCCGACGATC
## Slot "rightFixedSeq": TGGAATTCTCGGGTGCCAAGG
## Slot "leftFixedSeqOverlap": 5
## Slot "rightFixedSeqOverlap": 5
## Slot "confidenceLevel": 0.99
## Slot "minAffinity": 0.01
## Slot "missingValueSuppression": 0.5
## Slot "minSeedValue": 0.01
## Slot "seedLen": 15
## Slot "consensusSeq": [AG]G[AT]ACA[ACGT][ACGT][ACGT]TGT[AT]C[CT]
## Slot "upFootprintExtend": 4
## Slot "downFootprintExtend": 4
## Slot "fpLen": 23
##
## Fits a model of footprint length 23 for mono-nucleotide and shape features (shape = HelT, MGW) with fixed values for shape parameter positions not included in Shape.set used as offsets for the glm fit, and 11 view(s) per strand of DNA and 1 round(s) of data (round = 8) with reverse complement symmetry.
##
## Slot "regressionFormula": ObservedCount ~ offset(logProb)+offset(fixedSddG)+Round.8+N.A1+N.C1+N.G1+N.T1+N.A2+N.C2+N.G2+N.T2+N.A3+N.C3+N.G3+N.T3+N.A4+N.C4+N.G4+N.T4+N.A5+N.C5+N.G5+N.T5+N.A6+N.C6+N.G6+N.T6+N.A7+N.C7+N.G7+N.T7+N.A8+N.C8+N.G8+N.T8+N.A9+N.C9+N.G9+N.T9+N.A10+N.C10+N.G10+N.T10+N.A11+N.C11+N.G11+N.T11+N.A12+N.C12
##
## Slot "shapeParamsUsed[[1]]": HelT MGW
##
## Includes the following feature sub-classes:
## An object of class 'N'
## Fits 23 nucleotides for a feature model of length 23.
## Nucleotide features are reverse complement symmetric.
## Nucleotide beta values:
## 1 2 3 4 5
## N.A 0.006038364 0.00000000 0.00000000 0.00000000 0.00000000
## N.C 0.000000000 -0.05279580 -0.05543861 -0.14516828 -0.42643510
## N.G 0.007217709 0.02398093 -0.14383314 -0.06266071 -0.09687665
## N.T -0.036285376 -0.03426805 -0.03945558 -0.11732747 -0.60964801
## 6 7 8 9 10 11
## N.A -0.8004890 -0.8077337 0.0000000 -2.0071708 0.0000000 -0.34979259
## N.C -0.7779117 -1.1122550 -1.2187024 0.0000000 -1.0509835 0.00000000
## N.G 0.0000000 -0.2805733 -1.7420470 0.7441532 -0.5121767 -0.28601188
## N.T -0.8018140 0.0000000 -0.6716346 -1.9879509 -0.6155596 -0.08441138
## 12 13 14 15 16 17
## N.A -0.1057464 -0.08441138 -0.6155596 -1.9879509 -0.6716346 0.0000000
## N.C 0.0000000 -0.28601188 -0.5121767 0.7441532 -1.7420470 -0.2805733
## N.G 0.0000000 0.00000000 -1.0509835 0.0000000 -1.2187024 -1.1122550
## N.T -0.1057464 -0.34979259 0.0000000 -2.0071708 0.0000000 -0.8077337
## 18 19 20 21 22
## N.A -0.8018140 -0.60964801 -0.11732747 -0.03945558 -0.03426805
## N.C 0.0000000 -0.09687665 -0.06266071 -0.14383314 0.02398093
## N.G -0.7779117 -0.42643510 -0.14516828 -0.05543861 -0.05279580
## N.T -0.8004890 0.00000000 0.00000000 0.00000000 0.00000000
## 23
## N.A -0.036285376
## N.C 0.007217709
## N.G 0.000000000
## N.T 0.006038364
##
## Nucleotide beta errors:
## 1 2 3 4 5
## N.A 0.0001379786 0.0000000000 0.0000000000 0.0000000000 0.0000000000
## N.C 0.0000000000 0.0001561686 0.0001775221 0.0002587298 0.0010811718
## N.G 0.0001707361 0.0001454455 0.0002380502 0.0001628399 0.0001480179
## N.T 0.0001708647 0.0001757681 0.0001698314 0.0001657582 0.0006403368
## 6 7 8 9 10
## N.A 0.01356109 0.0001296038 0.00000000 0.03466908 0.0000000000
## N.C 0.03608456 0.0002964930 0.03644976 0.00000000 0.0416668328
## N.G 0.00000000 0.0004002780 0.02055038 0.02129473 0.0003196888
## N.T 0.00120320 0.0000000000 0.02059960 0.06596740 0.0050005605
## 11 12 13 14 15
## N.A 0.0004234784 0.0001785179 0.0001300629 0.0050005605 0.06596740
## N.C 0.0000000000 0.0000000000 0.0003276898 0.0003196888 0.02129473
## N.G 0.0003276898 0.0000000000 0.0000000000 0.0416668328 0.00000000
## N.T 0.0001300629 0.0001785179 0.0004234784 0.0000000000 0.03466908
## 16 17 18 19 20
## N.A 0.02059960 0.0000000000 0.00120320 0.0006403368 0.0001657582
## N.C 0.02055038 0.0004002780 0.00000000 0.0001480179 0.0001628399
## N.G 0.03644976 0.0002964930 0.03608456 0.0010811718 0.0002587298
## N.T 0.00000000 0.0001296038 0.01356109 0.0000000000 0.0000000000
## 21 22 23
## N.A 0.0001698314 0.0001757681 0.0001708647
## N.C 0.0002380502 0.0001454455 0.0001707361
## N.G 0.0001775221 0.0001561686 0.0000000000
## N.T 0.0000000000 0.0000000000 0.0001379786
##
##
## An object of class 'Intercept'
## Fits intercept(s) for 1 round(s) (round = 8).
## Intercept beta values:
## Round.8:
## [1] 39.37408
##
## Intercept beta errors:
## Round.8:
## [1] 0.002619789
##
##
##
## An object of class 'Shape'
## Fits 3 shape coefficients for 3 kinds of shape parameter(s) (shape = HelT, MGW) for a feature model of length 23.
## Shape features are reverse complement symmetric.
## Shape beta values:
## 1 2 3 4 5
## Shape.HelTA 0.012777702 0.009207855 0.01059514 0.02979068 -0.050599877
## Shape.HelTB 0.005339387 -0.003648404 -0.03570894 -0.01147808 -0.008140565
## Shape.MGW 0.011501160 -0.008920502 -0.02962055 -0.18939069 -0.087382882
## 6 7 8 9 10
## Shape.HelTA -0.04637238 -0.1065263 0.1929080 0.19583385 0.11049638
## Shape.HelTB -0.12908078 0.3710017 0.7437455 -0.03934332 -0.07502132
## Shape.MGW 0.06924589 0.3766392 1.2499150 0.17179767 -0.17550875
## 11 12 13 14 15
## Shape.HelTA 0.09218623 0.02171862 -0.03883268 -0.07502132 -0.03934332
## Shape.HelTB -0.03883268 0.02171862 0.09218623 0.11049638 0.19583385
## Shape.MGW -0.12381238 0.15027160 -0.12381238 -0.17550875 0.17179767
## 16 17 18 19 20
## Shape.HelTA 0.7437455 0.3710017 -0.12908078 -0.008140565 -0.01147808
## Shape.HelTB 0.1929080 -0.1065263 -0.04637238 -0.050599877 0.02979068
## Shape.MGW 1.2499150 0.3766392 0.06924589 -0.087382882 -0.18939069
## 21 22 23
## Shape.HelTA -0.03570894 -0.003648404 0.005339387
## Shape.HelTB 0.01059514 0.009207855 0.012777702
## Shape.MGW -0.02962055 -0.008920502 0.011501160
##
## Shape beta errors:
## 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23
## Shape.HelTA 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0
## Shape.HelTB 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0
## Shape.MGW 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0
##
## [1] "Number of Observations in Design Matrix: 986603"
## No shape parameters included in fit.
## [1] "i = 8"
## [1] "Round summary: "
## 8 Total
## Round 986603 986603
## [1] "Mono-nucleotide summary: "
## N.A N.C N.G N.T
## 1 575309 631680 343741 422476
## 2 708614 465992 367793 430807
## 3 901237 493080 165580 413309
## 4 917841 250637 403534 401194
## 5 1278078 17287 637145 40696
## 6 645 249 1958027 14285
## 7 828914 104025 92376 947891
## 8 1971244 312 912 738
## 9 570 1971847 428 361
## 10 1856339 681 113298 2888
## 11 77913 1099619 110817 684857
## 12 340280 646323 0 0
## [1] "View/strand orientation summary: "
## View.1 View.2 View.3 View.4 View.5 View.6 View.7 View.8 View.9
## Strand.F 5030 63192 95778 69057 106009 128500 139431 151738 110026
## Strand.R 0 0 0 0 0 0 0 0 0
## View.10 View.11 StrandTotal
## Strand.F 56300 61542 986603
## Strand.R 0 0 0
## [1] "Regression Formula: "
## [1] "ObservedCount ~ offset(logProb)+offset(fixedSddG)+N.A1+N.G1+N.T1+N.C2+N.G2+N.T2+N.C3+N.G3+N.T3+N.C4+N.G4+N.T4+N.C5+N.G5+N.T5+N.A6+N.C6+N.T6+N.A7+N.C7+N.G7+N.C8+N.G8+N.T8+N.A9+N.G9+N.T9+N.C10+N.G10+N.T10+N.A11+N.G11+N.T11+N.A12"
## Warning: glm.fit: fitted rates numerically 0 occurred
## No shape parameters included in fit.
## An object of class 'model'
##
## Slot "name": AR-DBD R8 Nucleotides + Fixed Shape Values (Rev. Comp. Sym.)
## Slot "varRegLen": 23
## Slot "leftFixedSeq": GTTCAGAGTTCTACAGTCCGACGATC
## Slot "rightFixedSeq": TGGAATTCTCGGGTGCCAAGG
## Slot "leftFixedSeqOverlap": 5
## Slot "rightFixedSeqOverlap": 5
## Slot "confidenceLevel": 0.99
## Slot "minAffinity": 0.01
## Slot "missingValueSuppression": 0.5
## Slot "minSeedValue": 0.01
## Slot "seedLen": 15
## Slot "consensusSeq": [AG]G[AT]ACA[ACGT][ACGT][ACGT]TGT[AT]C[CT]
## Slot "upFootprintExtend": 4
## Slot "downFootprintExtend": 4
## Slot "fpLen": 23
##
## Fits a model of footprint length 23 for mono-nucleotide and shape features (shape = HelT, MGW) with fixed values for shape parameter positions not included in Shape.set used as offsets for the glm fit, and 11 view(s) per strand of DNA and 1 round(s) of data (round = 8) with reverse complement symmetry.
##
## Slot "regressionFormula": ObservedCount ~ offset(logProb)+offset(fixedSddG)+Round.8+N.A1+N.C1+N.G1+N.T1+N.A2+N.C2+N.G2+N.T2+N.A3+N.C3+N.G3+N.T3+N.A4+N.C4+N.G4+N.T4+N.A5+N.C5+N.G5+N.T5+N.A6+N.C6+N.G6+N.T6+N.A7+N.C7+N.G7+N.T7+N.A8+N.C8+N.G8+N.T8+N.A9+N.C9+N.G9+N.T9+N.A10+N.C10+N.G10+N.T10+N.A11+N.C11+N.G11+N.T11+N.A12+N.C12
##
## Slot "shapeParamsUsed[[1]]": HelT MGW
##
## Includes the following feature sub-classes:
## An object of class 'N'
## Fits 23 nucleotides for a feature model of length 23.
## Nucleotide features are reverse complement symmetric.
## Nucleotide beta values:
## 1 2 3 4 5
## N.A 0.006038310 0.00000000 0.00000000 0.00000000 0.00000000
## N.C 0.000000000 -0.05279567 -0.05543868 -0.14516820 -0.42643505
## N.G 0.007217837 0.02398088 -0.14383322 -0.06266048 -0.09687667
## N.T -0.036285438 -0.03426806 -0.03945569 -0.11732744 -0.60964800
## 6 7 8 9 10 11
## N.A -0.8004890 -0.8077336 0.0000000 -2.0071709 0.0000000 -0.34979092
## N.C -0.7779117 -1.1122550 -1.2187028 0.0000000 -1.0509835 0.00000000
## N.G 0.0000000 -0.2805732 -1.7420469 0.7441527 -0.5121768 -0.28601192
## N.T -0.8018141 0.0000000 -0.6716349 -1.9879514 -0.6155596 -0.08441141
## 12 13 14 15 16 17
## N.A -0.1057464 -0.08441141 -0.6155596 -1.9879514 -0.6716349 0.0000000
## N.C 0.0000000 -0.28601192 -0.5121768 0.7441527 -1.7420469 -0.2805732
## N.G 0.0000000 0.00000000 -1.0509835 0.0000000 -1.2187028 -1.1122550
## N.T -0.1057464 -0.34979092 0.0000000 -2.0071709 0.0000000 -0.8077336
## 18 19 20 21 22
## N.A -0.8018141 -0.60964800 -0.11732744 -0.03945569 -0.03426806
## N.C 0.0000000 -0.09687667 -0.06266048 -0.14383322 0.02398088
## N.G -0.7779117 -0.42643505 -0.14516820 -0.05543868 -0.05279567
## N.T -0.8004890 0.00000000 0.00000000 0.00000000 0.00000000
## 23
## N.A -0.036285438
## N.C 0.007217837
## N.G 0.000000000
## N.T 0.006038310
##
## Nucleotide beta errors:
## 1 2 3 4 5
## N.A 0.0001379787 0.0000000000 0.0000000000 0.0000000000 0.0000000000
## N.C 0.0000000000 0.0001561686 0.0001775221 0.0002587298 0.0010811718
## N.G 0.0001707361 0.0001454455 0.0002380502 0.0001628401 0.0001480178
## N.T 0.0001708647 0.0001757681 0.0001698315 0.0001657582 0.0006403368
## 6 7 8 9 10
## N.A 0.01356109 0.0001296039 0.00000000 0.03466908 0.0000000000
## N.C 0.03608456 0.0002964930 0.03644978 0.00000000 0.0416668328
## N.G 0.00000000 0.0004002780 0.02055038 0.02129476 0.0003196888
## N.T 0.00120320 0.0000000000 0.02059962 0.06596736 0.0050005605
## 11 12 13 14 15
## N.A 0.0004234824 0.000178518 0.0001300629 0.0050005605 0.06596736
## N.C 0.0000000000 0.000000000 0.0003276898 0.0003196888 0.02129476
## N.G 0.0003276898 0.000000000 0.0000000000 0.0416668328 0.00000000
## N.T 0.0001300629 0.000178518 0.0004234824 0.0000000000 0.03466908
## 16 17 18 19 20
## N.A 0.02059962 0.0000000000 0.00120320 0.0006403368 0.0001657582
## N.C 0.02055038 0.0004002780 0.00000000 0.0001480178 0.0001628401
## N.G 0.03644978 0.0002964930 0.03608456 0.0010811718 0.0002587298
## N.T 0.00000000 0.0001296039 0.01356109 0.0000000000 0.0000000000
## 21 22 23
## N.A 0.0001698315 0.0001757681 0.0001708647
## N.C 0.0002380502 0.0001454455 0.0001707361
## N.G 0.0001775221 0.0001561686 0.0000000000
## N.T 0.0000000000 0.0000000000 0.0001379787
##
##
## An object of class 'Intercept'
## Fits intercept(s) for 1 round(s) (round = 8).
## Intercept beta values:
## Round.8:
## [1] 39.37408
##
## Intercept beta errors:
## Round.8:
## [1] 0.002619789
##
##
##
## An object of class 'Shape'
## Fits 3 shape coefficients for 3 kinds of shape parameter(s) (shape = HelT, MGW) for a feature model of length 23.
## Shape features are reverse complement symmetric.
## Shape beta values:
## 1 2 3 4 5
## Shape.HelTA 0.012777702 0.009207855 0.01059514 0.02979068 -0.050599877
## Shape.HelTB 0.005339387 -0.003648404 -0.03570894 -0.01147808 -0.008140565
## Shape.MGW 0.011501160 -0.008920502 -0.02962055 -0.18939069 -0.087382882
## 6 7 8 9 10
## Shape.HelTA -0.04637238 -0.1065263 0.1929080 0.19583385 0.11049638
## Shape.HelTB -0.12908078 0.3710017 0.7437455 -0.03934332 -0.07502132
## Shape.MGW 0.06924589 0.3766392 1.2499150 0.17179767 -0.17550875
## 11 12 13 14 15
## Shape.HelTA 0.09218623 0.02171862 -0.03883268 -0.07502132 -0.03934332
## Shape.HelTB -0.03883268 0.02171862 0.09218623 0.11049638 0.19583385
## Shape.MGW -0.12381238 0.15027160 -0.12381238 -0.17550875 0.17179767
## 16 17 18 19 20
## Shape.HelTA 0.7437455 0.3710017 -0.12908078 -0.008140565 -0.01147808
## Shape.HelTB 0.1929080 -0.1065263 -0.04637238 -0.050599877 0.02979068
## Shape.MGW 1.2499150 0.3766392 0.06924589 -0.087382882 -0.18939069
## 21 22 23
## Shape.HelTA -0.03570894 -0.003648404 0.005339387
## Shape.HelTB 0.01059514 0.009207855 0.012777702
## Shape.MGW -0.02962055 -0.008920502 0.011501160
##
## Shape beta errors:
## 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23
## Shape.HelTA 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0
## Shape.HelTB 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0
## Shape.MGW 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0
##
## [1] "Number of Observations in Design Matrix: 986603"
## No shape parameters included in fit.
## [1] "Stability Reached after 8 iterations."
ModelTest <- finalizeFeatureBetas(ModelTest)
pM <- plot(ModelTest, plotTitle = "AR-DBD R8 Nucleotides + Fixed Shape Values Fit", Nplot.ddG = TRUE, verticalPlots = TRUE)
ggplot2::ggsave(pM, file = paste(selexDir, saveDir, "/modelPlot.pdf", sep = ""), height = vPheight, width = 6)
save(ModelTest, file = paste(selexDir, saveDir, "/model.RData",sep = ""))
saveRDS(ModelTest, file = paste(selexDir, saveDir, "/model.rds",sep = ""))