Peak Counting Score

The Peak Counting Score (abbreviation PCS ) is a value that indicates the similarity of two mass spectra . The PCS depends on the number of common and different peak values ​​( peaks ) in both spectra. The value is a very simple way of comparing two mass spectra with one another.

calculation

Given are:

• S A  : The mass spectrum A as a peak list
• S B  : The mass spectrum B as a peak list
• δ  : A threshold value that indicates below which mass difference 2 peaks should be considered similar
• ε  : The empty peak (if a peak only occurs in one spectrum, ε stands for the position in the other spectrum)

${\displaystyle pcs(S_{A},S_{B})=\sum _{\rm {i\,{\ddot {a}}hnlich\,j}}score(i,j)+\sum _{\text{i nicht in B}}score(i,{\rm {\varepsilon }})+\sum _{\text{j nicht in A}}score({\rm {\varepsilon }},j)}$

${\displaystyle score(i,j)={\begin{cases}1,&{\text{wenn }}\left|masse(i)-masse(j)\right|\leq {\rm {\delta }}\\0,&{\text{sonst}}\end{cases}}}$

${\displaystyle score(i,{\rm {\varepsilon }})=score({\rm {\varepsilon }},j)=0}$

example

• S A = {10,20,30,50}
• S B = {22.30,45,53}
• δ = 4

The alignment in this case would be:

Peak list A	10	20th	30th	ε	50
Peak list B	ε	22nd	30th	45	53

The value pairs 20 and 22, 30 and 30 or 50 and 53 are classified as hits ( match ), since their difference amounts are less than or equal to δ = 4.

The PCS would therefore be here: ${\displaystyle {\begin{aligned}pcs(S_{A},S_{B})&=score(10,{\rm {\varepsilon }})+score(20,22)+score(30,30)+score({\rm {\varepsilon }},45)+score(50,53)\\&=0+1+1+0+1\\&=3\end{aligned}}}$

Web links

• Technical article on the subject of IT methods for protein identification
• Slides on the topic (PDF file; 373 kB)