Scripture doesn't recognize CIGAR string.

[thejustpark]

Let me describe my problem with an example.

This is my sorted, indexed sam file.
seq.2851795 89 1 4470815 255 49M98N29M2S = 4470815 0 TGTGTGTGTGTGTGCGCGCGCGTGTGTGTGTGTGTGTGTGTGTGTGTGTGTGTGTGTGTGTGTGTATGTGTATGTGTGCG @@@DD@DDGDDFHGFHHGGIIEGGADBHIIIIIDEIIIHIIJGHEEGHGGHFEFDCC18/=/;;@)0&8>?@>B&25<98 OL:A:IH:i:1 HI:i:1

When I run Scripture on this with appropriate size information and reference genome, I got this error.

Using Version VPaperR3
Has pairs: false
Has upweighting turned on: false
Computing weights..... upweighting? false weight: 1.0
AlignmentDataModel loaded, initializing model stats
Computing alignment global stats for chromosome 1
model stats loaded, initializing model
Built the model: 0.0010 free memory: 117707336
Loaded chromosome Sequence
Segmenting accross graph
Going to get read iterator to make graph with counts
Got read iterator
Made it through all reads
Collapsed reads
Decollapsed by introns
Made first graph
Got extended pieces
Exception in thread "main" java.lang.NullPointerException
at broad.pda.seq.segmentation.ContinuousDataAlignmentModel.getNodeCounts(ContinuousDataAlignmentModel.java:1395)
at broad.pda.seq.segmentation.ContinuousDataAlignmentModel.makeGraphWithCounts(ContinuousDataAlignmentModel.java:1359)
at broad.pda.seq.segmentation.ContinuousDataAlignmentModel.contiguousAcrossGraph(ContinuousDataAlignmentModel.java:1192)
at broad.pda.seq.segmentation.ContinuousDataAlignmentModel.main(ContinuousDataAlignmentModel.java:2209)

But when I change the 5th column (the CIGAR string) into 80M, it works and returns a result.
Scripture's web page says it takes in a `spliced read alignment', and my input indicates that it is a spliced read alignment. Then what's wrong in my file?

FYI, I used both of the working version of Scripture (scripture-beta.jar and scripture-beta2.jar)
Thanks in advance,
HJ.