@@ -1,8 +1,8 @@

 Package: MotifDb

 Type: Package

 Title: An Annotated Collection of Protein-DNA Binding Sequence Motifs

-Version: 1.1.0

-Date: 2012-08-12

+Version: 1.1.2

+Date: 2012-10-29

 Author: Paul Shannon

 Maintainer: Paul Shannon <pshannon@fhcrc.org>

 Depends: R (>= 2.15.0), methods, IRanges, Biostrings

new file mode 100644

@@ -0,0 +1,7 @@

+CHANGES IN VERSION 1.1.2

+-------------------------

+

+NEW FEATURES

+

+    o 683 new motifs derived from DGF (digital genome footprinting) 

+      from the stamlab (http://www.stamlab.org/) added.

@@ -1,36 +1,36 @@

 MotifDb <- NULL

 #-------------------------------------------------------------------------------

-.MotifDb = function (loadAllSources=TRUE, quiet=TRUE)

+.MotifDb = function(loadAllSources=TRUE, quiet=TRUE)

 {

-  mdb = MotifList ()

+  mdb = MotifList()

-  if (loadAllSources) {

-    data.path = system.file ('extdata', package='MotifDb') # e (system.file (package='MotifDb'), 'data', sep='/')

-    data.files = dir (data.path, full.names=TRUE)

+  if(loadAllSources) {

+    data.path = system.file('extdata', package='MotifDb')

+    data.files = dir(data.path, full.names=TRUE)

-    if (length (data.files) > 0)   

-      for (data.file in data.files) {

+    if(length(data.files) > 0)   

+      for(data.file in data.files) {

          # define these to keep 'check' happy.  they are loaded by 'load'

         tbl.md = NA; matrices = NA;  

-        variables = load (data.file)

-        mdb = append (mdb, MotifList (matrices, tbl.md))

-        if (!quiet) 

-          message (noquote (sprintf ('added %s (%d) matrices, length now: %d',

-                   basename (data.file), length (matrices), length (mdb))))

+        variables = load(data.file)

+        mdb = append(mdb, MotifList(matrices, tbl.md))

+        if(!quiet) 

+          message(noquote(sprintf('added %s(%d) matrices, length now: %d',

+                   basename(data.file), length(matrices), length(mdb))))

       } # for data.file

-    if (!quiet) {

-      print (table (values (mdb)$dataSource))

+    if(!quiet) {

+      print(table(values(mdb)$dataSource))

       }

     } # if loadAllSources

-  return (mdb)

+  return(mdb)

 } # MotifDb

 #-------------------------------------------------------------------------------

 .onLoad <- function(libname, pkgname)

 {

-    MotifDb <<- .MotifDb (loadAllSources=TRUE, quiet=TRUE)

+    MotifDb <<- .MotifDb(loadAllSources=TRUE, quiet=TRUE)

 }

 #-------------------------------------------------------------------------------

new file mode 100644

Binary files /dev/null and b/inst/extdata/stamlab.RData differ

@@ -121,12 +121,10 @@ test.allMatricesAreNormalized = function ()

   print ('--- test.allMatricesAreNormalized')

   mdb = MotifDb# (quiet=TRUE)

   matrices = mdb@listData

-  colsums = as.integer (sapply (matrices, function (mtx) as.integer (mean (round (colSums (mtx))))))

-  failures = which (colsums != 1)

-  if (length (failures > 0))

-    browser ()

-  checkTrue (length (failures) == 0)

-

+    # a lenient test required by "Cparvum-UniPROBE-Cgd2_3490.UP00395" and  "Hsapiens-UniPROBE-Sox4.UP00401"

+    # for reasons not yet explored.  10e-8 should be be possible

+  checkTrue (all (sapply (matrices, function (m) all (abs (colSums (m) - 1.0) < 0.02))))

+             

 } # test.allMatricesAreNormalized

 #------------------------------------------------------------------------------------------------------------------------

 test.providerNames = function ()

@@ -144,7 +142,7 @@ test.geneSymbols = function ()

   print ('--- test.getGeneSymbols')

   mdb = MotifDb # ()

   syms = values (mdb)$geneSymbol

-  checkEquals (length (which (is.na (syms))), 0)

+  checkEquals (length (which (is.na (syms))), 683)  # no symols yet for the dgf stamlab motifs

   checkEquals (length (which (syms == '')), 0)

 } # test.geneSymbols

@@ -157,10 +155,10 @@ test.geneIdsAndTypes = function ()

   geneIdTypes = values (mdb)$geneIdType

   tbl.types = as.data.frame (table (geneIdTypes, useNA='always'), stringsAsFactors=FALSE)

   checkEquals (tbl.types$geneIdTypes,  c ('ENTREZ', 'FLYBASE', 'SGD', NA))

-  checkEquals (tbl.types$Freq,  c (763, 614, 453, 256))

+  checkEquals (tbl.types$Freq,  c (763, 614, 453, 939))

   na.count = length (which (is.na (geneIds)))

-  checkEquals (na.count, 256)   # see geneIdTypes == NA, just above

+  checkEquals (na.count, 939)   # see geneIdTypes == NA, just above

   empty.count = length (which (geneIds == ''))

   checkEquals (empty.count, 0)

@@ -21,13 +21,14 @@ FlyFactorSurvey: \tab 614\cr

 hPDI: \tab 437\cr

 JASPAR_CORE: \tab 459\cr

 ScerTF: \tab 196\cr

+stamlab: \tab 683\cr

 UniPROBE: \tab 380\cr

 }

 Representing primarily four organsisms:

 \tabular{ll}{

 Dmelanogaster: \tab 739\cr

-Hsapiens: \tab 505\cr

+Hsapiens: \tab 1188\cr

 Scerevisiae: \tab 464\cr

 Mmusculus: \tab 329\cr

 Rnorvegicus: \tab 8\cr

@@ -172,6 +173,10 @@ an example (see below for the associated position frequency matrix):

 \itemize{

+\item Neph S, Stergachis AB, Reynolds A, Sandstrom R, Borenstein E,

+Stamatoyannopoulos JA. Circuitry and dynamics of human transcription factor regulatory networks.

+Cell. 2012 Sep 14;150(6):1274-86.

+  

 \item Portales-Casamar E, Thongjuea S, Kwon AT, Arenillas D, Zhao X, Valen E, Yusuf D, Lenhard B, Wasserman WW, Sandelin A. JASPAR 2010: the greatly expanded open-access database of transcription factor binding profiles. Nucleic Acids Res. 2010 Jan;38(Database issue):D105-10. Epub 2009 Nov 11.

 \item Robasky K, Bulyk ML. UniPROBE, update 2011: expanded content and search tools in the online database of protein-binding microarray data on protein-DNA interactions. Nucleic Acids Res. 2011 Jan;39(Database issue):D124-8. Epub 2010 Oct 30.