[MINC-users] calculating r-squared values
Jared Rowley
jared.rowley at gmail.com
Mon Jun 11 11:10:26 EDT 2012
Got it. That makes sense.
Thanks very much for the help,
Jared
On Mon, Jun 11, 2012 at 11:07 AM, Jason Lerch <jason at phenogenomics.ca>wrote:
> You have to save your output to a variable:
>
> output <- mincApply(gf1$FA, quote(fff(x)))
>
> and then call mincWriteVolume(output, "filename.mnc")
>
> If you didn't do that, the internal R variable .Last.value would have
> saved your last computation - though if you did anything else since then
> (and the mincWriteVolume call might apply, I'm not sure), then you'll have
> to rerun mincApply and assign its output to some variable.
>
> Jason
>
> On 2012-06-11, at 10:25 AM, Jared Rowley wrote:
>
> > Hi Jason,
> >
> > Everything appeared to work however I'm a little unclear how to save the
> > data now. (It does take a very long time to run).
> >
> > Here is my code. The last line was my attempt to save the data:
> >
> > gf1 <- read.csv("dti_nvm_gd.csv")
> >> fff <- function(x) { return(summary(lm(x ~ score, gf1))$r.squared) }
> >> mincApply(gf1$FA, quote(fff(x)))
> > Number of volumes: 91
> > Volume sizes: 193 229 193
> > In slice
> > 0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20
> > 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38
> 39
> > 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57
> 58
> > 59 60 61 62 63 64 65 66 67 68 69 70 71 72 73 74 75 76
> 77
> > 78 79 80 81 82 83 84 85 86 87 88 89 90 91 92 93 94 95
> 96
> > 97 98 99 100 101 102 103 104 105 106 107 108 109 110 111
> > 112 113 114 115 116 117 118 119 120 121 122 123 124 125 126
> > 127 128 129 130 131 132 133 134 135 136 137 138 139 140 141
> > 142 143 144 145 146 147 148 149 150 151 152 153 154 155 156
> > 157 158 159 160 161 162 163 164 165 166 167 168 169 170 171
> > 172 173 174 175 176 177 178 179 180 181 182 183 184 185 186
> > 187 188 189 190 191 192
> > Done
> >
> >> mincWriteVolume(fff,"rsquared_test.mnc", "score")
> > Error in minc.dimensions.sizes(like.filename) :
> > Error opening input file: score.
> >
> > Thanks again,
> >
> > Jared
> >
> >
> > On Fri, Jun 8, 2012 at 12:39 PM, Jason Lerch <jason at phenogenomics.ca>
> wrote:
> >
> >> Try replacing FA with x - mincApply assigns the current voxel to the
> >> variable x in its loop. I.e.:
> >>
> >> library(RMINC)
> >> gf1 <- read.csv("dti_gd.csv")
> >> fff <- function(x) { return(summary(lm(x ~ score, gf1))$r.squared) }
> >> mincApply(gf1$FA, quote(fff(x)))
> >>
> >> Jason
> >>
> >> On 2012-06-08, at 11:32 AM, Jared Rowley wrote:
> >>
> >>> Hi Jason,
> >>>
> >>> Thanks for the response.
> >>>
> >>> I did the following and got an error do you have any suggestions:
> >>>
> >>> rm(list=ls(all=TRUE))
> >>>> library(RMINC)
> >>>> gf1 <- read.csv("dti_gd.csv")
> >>>> fff <- function(x) { return(summary(lm(FA ~ score, gf1))$r.squared) }
> >>>> mincApply(gf1$FA, quote(fff(x)))
> >>> Number of volumes: 91
> >>> Volume sizes: 193 229 193
> >>> In slice
> >>> 0 Error in storage.mode(y) <- "double" :
> >>> invalid to change the storage mode of a factor
> >>> In addition: Warning message:
> >>> In model.response(mf, "numeric") :
> >>> using type="numeric" with a factor response will be ignored
> >>>
> >>> FA is a column of minc files and score is a column of integers I want
> to
> >>> correlate with FA.
> >>>
> >>> Thanks very much,
> >>>
> >>> Jared
> >>>
> >>>
> >>> On Wed, Jun 6, 2012 at 1:35 PM, Jason Lerch <jason at phenogenomics.ca>
> >> wrote:
> >>>
> >>>> This should work, though is going to be slow:
> >>>>
> >>>> fff <- function(x) { return(summary(lm(x ~ genotype, gf))$r.squared) }
> >>>>
> >>>> mincApply(filenames, quote(fff(x)))
> >>>>
> >>>> Note: replace the formula in that function with the appropriate
> formula
> >>>> for your purposes, and filenames should be a vector of your filenames.
> >>>> Could be worth specifying a mask, as this will run for a few hours.
> >>>>
> >>>> And I haven't actually tested it, so it might not run at all!
> >>>>
> >>>> Jason
> >>>>
> >>>> On 2012-06-05, at 4:18 PM, Jared Rowley wrote:
> >>>>
> >>>>> Hi Minc Users,
> >>>>>
> >>>>> I was wondering if anyone knows a way calculate R-squared values on a
> >>>> voxel
> >>>>> level from a linear regression.
> >>>>>
> >>>>> I assume there is a way with either glim_image or RMINC however
> reading
> >>>>> though the documentation I could not find a solution.
> >>>>>
> >>>>> Thanks very much,
> >>>>>
> >>>>> Jared
> >>>>> _______________________________________________
> >>>>> MINC-users at bic.mni.mcgill.ca
> >>>>> http://www.bic.mni.mcgill.ca/mailman/listinfo/minc-users
> >>>>
> >>>> _______________________________________________
> >>>> MINC-users at bic.mni.mcgill.ca
> >>>> http://www.bic.mni.mcgill.ca/mailman/listinfo/minc-users
> >>>>
> >>>
> >>>
> >>>
> >>> --
> >>> Jared Rowley
> >>> Translational Neuroimaging Laboratory
> >>> McGill Centre for Studies in Aging
> >>> Douglas Hospital Research Center
> >>> McGill University
> >>> 6825 LaSalle Blvd.
> >>> Montreal, QC Canada
> >>> H4H 1R3
> >>> _______________________________________________
> >>> MINC-users at bic.mni.mcgill.ca
> >>> http://www.bic.mni.mcgill.ca/mailman/listinfo/minc-users
> >>
> >> _______________________________________________
> >> MINC-users at bic.mni.mcgill.ca
> >> http://www.bic.mni.mcgill.ca/mailman/listinfo/minc-users
> >>
> >
> >
> >
> > --
> > Jared Rowley
> > Translational Neuroimaging Laboratory
> > McGill Centre for Studies in Aging
> > Douglas Hospital Research Center
> > McGill University
> > 6825 LaSalle Blvd.
> > Montreal, QC Canada
> > H4H 1R3
> > _______________________________________________
> > MINC-users at bic.mni.mcgill.ca
> > http://www.bic.mni.mcgill.ca/mailman/listinfo/minc-users
>
> _______________________________________________
> MINC-users at bic.mni.mcgill.ca
> http://www.bic.mni.mcgill.ca/mailman/listinfo/minc-users
>
--
Jared Rowley
Translational Neuroimaging Laboratory
McGill Centre for Studies in Aging
Douglas Hospital Research Center
McGill University
6825 LaSalle Blvd.
Montreal, QC Canada
H4H 1R3
More information about the MINC-users
mailing list