Nadasdy's Work Page
PROJECT:
REPEATING SPATIO-TEMPORAL PATTERNS OF NEURONAL ACTIVITY IN THE HIPPOCAMPUS DURING SLEEP.
DATA DESCRIPTION
8 neurons (putative: 7 pyramidal and 1 interneuron)
50 microsec sampling rate
mostly (SPW)
 parspace_seq_diff_dt.gif
Parameterspace: different dT tolerances
|
#shuffl:no shuffl
#search:seq
#repeat:1-20
#ref: 6
#seq:2
#twin:4000
#toler: var
#outp: param_ch_sq_$k.dat
for k in 50 100 150 200 250 300
do
rm param_ch_sq_$k.dat
for i in `count 1 40`
do
echo $k $i
rm scroll.???
rm vectors.???
#echo $j >> shuffled_ch_sq_$k.dat
#sh seq_shuffl_nut.ksh 8000 230 $i 6 $k
ptrn_search_gold 8 junk.00B junk.00D junk.00F junk.0BA junk.00C junk.00E junk.00G junk.0AA junk 2 $i 6 4000 $k >> param_ch_sq_$k.dat
done
more CR >> param_ch_sq_$k.dat
done
|
|---|
Monte Carlo Simulation of Pattern Detection
In the following simulations spikes were moved to a randomly selected different cell (channel). Complex spikes were treated as single events. Therefore, the entire complex burst was moved to the same channel. This way the complex spike integrity was preserved but the sequential relationship between single or complex spikes was eliminated. The synchronous and intermittent frequency modulation of the spiketrains was also preserved. The tolerance (dT) was varied between 5 msec to 30 msec (from 50 points to 300 points). Sequences were detected by the 'Sequence searching algorithm'. In the figures we compared the number of sequences and the number of repetition between the 20 randomized and the one original database.
shell for the re00.sm.gif">
param_ch_sq_300.gif
|
shell for the real data:
#shuffl:no shuffl
#search:seq
#repeat:1-20
#ref: 6
#seq:2
#twin:4000
#toler: var
#outp: param_ch_sq_$k.dat
for k in 50 100 150 200 250 300
do
rm param_ch_sq_$k.dat
echo $j
for i in `count 1 20`
do
#echo $j $i
rm scroll.???
rm vectors.???
#echo $j >> shuffled_ch_sq_$k.dat
#sh seq_shuffl_nut.ksh 8000 230 $i 6 $k
ptrn_search_gold 8 junk.00B junk.00D junk.00F junk.0BA junk.00C junk.00E junk.00G junk.0AA junk 2 $i 6 4000 $k >> param_ch_sq_$k.dat
done
more CR >> param_ch_sq_$k.dat
done
Shell for the shuffled data:
#shuffl:ch
#search:seq
#repeat:1-20
#ref: 6
#seq:2
#twin:4000
#toler: var
#outp: shuffled_ch_sq_$k.dat
for k in 50 100 150 200 250 300
do
rm shuffled_ch_sq_$k.dat
for j in `count 1 20`
do
shuffling_chcpx 8 junk.00B junk.00D junk.00F junk.0BA junk.00C junk.00E junk.00G junk.0AA 140
echo $j
for i in `count 1 20`
do
#echo $j $i
rm scroll.???
rm vectors.???
#echo $j >> shuffled_ch_sq_$k.dat
#sh seq_shuffl_nut.ksh 8000 230 $i 6 $k
ptrn_search_gold 8 junk.00B.shf junk.00D.shf junk.00F.shf junk.0BA.shf junk.00C.shf junk.00E.shf junk.00G.shf junk.0AA.shf junk 2 $i 6 4000 $k >> shuffled_ch_sq_$k.dat
done
more CR >> shuffled_c00 $k >> shuffled_ch_sq_$k.dat
done
more CR >> shuffled_ch_sq_$k.dat
done
done
DATA DESCRIPTION
5 neurons (putative: 5 pyramidal cells)
100 microsec sampling rate
mostly (SPW and theta)
Datafile: j06-08
DATA DESCRIPTION
10 neurons (putative: 10 pyramidal cells)
50 microsec sampling rate
mostly (SPW and theta)
Datafile: k9-02
DATA DESCRIPTION
5 neurons (putative: 5 pyramidal cells)
50 microsec sampling rate
mostly (SPW and theta)
2 weel running cells
Datafile: k12-30
Animal: K19-18
7 pyramidal cells (clustr.0AC-clustr.0AK)
50 microsec sampling rate
Behav: wheel running (theta and non-theta)
Behav: some cells are wheel direction cells
Questions and comments: zoli@osiris.rutgers.edu
/h3>