I used findepeaks, using method "topology", for detecting peaks in unidimensional sequence data. The data is numeric, and I tried multiple limits to find a sensible value for the limit. I am trying to replicate the analysis I did a couple of years ago. It seems that the behaviour of "limit" has changed, and I cannot replicate the results. It shows strange behaviour. For instance, analysing seq 2, it cannot detect a peak, error at 3, but shows a peak at the limit of 5, 10, 15, or even 50. Should not the score at peak be lower than the limit? Here is the sequence of data (ordered)?
Here are two sample sequences (ordered frequencies).
Id Seq1. Seq2
0 1 0
1 1 0
2 0 0
3 1 0
4 2 0
5 1 0
6 2 0
7 3 0
8 5 0
9 3 0
10 4 0
11 1 0
12 2 0
13 7 0
14 5 0
15 5 0
16 8 0
17 6 0
18 5 0
19 7 0
20 6 1
21 13 1
22 7 0
23 10 1
24 14 0
25 7 0
26 11 0
27 5 1
28 8 0
29 12 0
30 10 3
31 5 0
32 7 0
I used findepeaks, using method "topology", for detecting peaks in unidimensional sequence data. The data is numeric, and I tried multiple limits to find a sensible value for the limit. I am trying to replicate the analysis I did a couple of years ago. It seems that the behaviour of "limit" has changed, and I cannot replicate the results. It shows strange behaviour. For instance, analysing seq 2, it cannot detect a peak, error at 3, but shows a peak at the limit of 5, 10, 15, or even 50. Should not the score at peak be lower than the limit? Here is the sequence of data (ordered)?
Here are two sample sequences (ordered frequencies).
Id Seq1. Seq2
0 1 0
1 1 0
2 0 0
3 1 0
4 2 0
5 1 0
6 2 0
7 3 0
8 5 0
9 3 0
10 4 0
11 1 0
12 2 0
13 7 0
14 5 0
15 5 0
16 8 0
17 6 0
18 5 0
19 7 0
20 6 1
21 13 1
22 7 0
23 10 1
24 14 0
25 7 0
26 11 0
27 5 1
28 8 0
29 12 0
30 10 3
31 5 0
32 7 0