[Gmod-gbrowse] problem in displaying individual paired-end reads on gbrowse instance

[Vipin TS]

Hello Gbrowse team,

Recently I noticed that, on our public Gbrowse instance (
http://gbrowse.cbio.mskcc.org/), the individual read display is not
properly rendering. (works fine for single-end reads)

The error message from apache log:

RenderPanels error: Can't locate object method "tid" via package
"Bio::DB::Bam::SplitAlignmentPart" at
/data/gb/perl5/lib/perl5/x86_64-linux-thread-multi//Bio/DB/Sam.pm line 1991.
[Wed Oct 01 11:08:28 2014] [warn] [client 172.21.50.98] mod_fcgid: stderr:
[19371] rendering error track: Can't locate object method "tid" via package
"Bio::DB::Bam::SplitAlignmentPart" at
/data/gb/perl5/lib/perl5/x86_64-linux-thread-multi//Bio/DB/Sam.pm line
1991. at
/data/gb/gbrowse/lib/perl5/x86_64-linux-thread-multi//Bio/Graphics/Browser2/Render.pm
line 3678., referer: http://gbrowse.cbio.mskcc.org/gb/gbrowse/human/

Here in my track configurations:
[TSkim:database]
db_adaptor = Bio::DB::Sam
db_args = -bam /data/gb/gbrowse/www/data/TSkim_uniq.bam
-split_splices 1

[TSkim_reads:1001]
feature = coverage
database = TSkim
glyph = wiggle_xyplot
bgcolor = red
height = 40
fgcolor = black
min_score = 0
scale = three
label = 0
category = 4. RNA-Sequencing (ARTS2 project)
key = SRR534327

[TSkim_reads]
feature = read_pair
database = TSkim
parts_overlap= 1
glyph = segments
draw_target = 1
show_mismatch = 1
mismatch_color = red
indel_color = green
height = 3
label = 0
label density = 50
fgcolor = black
bgcolor = green
maxdepth = 2
box_subparts = 2
stranded = 1
connector = dashed
bump = fast
category = 4. RNA-Sequencing (ARTS2 project)
key = SRR534327

The above configuration works fine for displaying the coverage tracks, but
when it comes to individual read display the error happens.

I am using Bio::DB::Sam version 1.38, from the
http://search.cpan.org/~lds/Bio-SamTools-1.38/ distribution package page, I
am seeing the "Bio::DB::Bam::SplitAlignmentPart " seems like a deprecated
module.

Can anyone please let me know is there any issue with the track
configuration? or a better way for displaying paired end reads.

Thank you,

Vipin | RÀtsch Lab