examples of mcr-als applications (from r.tauler et al.) - cid… · rank-deficient matrices a data...
TRANSCRIPT
Examples of MCR-ALS applications (from R.Tauler et al.)
•Chromatographic coelution•Spectrometric titrations of multiequilibriasystems•Process analysis• continuous flow methods FIA•melting conformations of polynucleotides•protein folding•voltamperometric data (Me-peptide interactions)•environmental data
R.Tauler, A.Izquierdo-Ridorsa and E.CasassasChemometrics and Intelligent Laboratory Systems, 18 (1993, 293-300
Spectrometric titrations: An easy way for the generation of two- and three-way data in the study of chemical reactions and interactions
T=37oC
Spectrophotometer
Peristalticpump
0.050 ml
-125.3
pHmeter
Autoburette
Computer
PrinterStirrer
Thermostatic bath
400 450 500 550 600 650 700 750 800 850 9000
0.1
0.2
0.3
0.4
400 450 500 550 600 650 700 750 800 850 9000
0.1
0.2
0.3
0.4
0.5
400 450 500 550 600 650 700 750 800 850 9000
0.1
0.2
0.3
0.4
0.5
n
Three spectrometric titrations of a multiequilibria system:Cu(II)-salicylate
3 4 5 6 7 8 90
10
20
30
40
50
60
70
80
90
100
pH
.- individual anal., +- simoult. anal., --co
ncen
tratio
n in
%
3 4 5 6 7 8 90
10
20
30
40
50
60
70
80
90
100
pH
conc
entra
tion
in %
.- individual anal., +- simoult. anal., --
3 4 5 6 7 8 90
10
20
30
40
50
60
70
80
90
100
pH
conc
entra
tion
in %
.- individual anal., +- simoult. anal., --
400 500 600 700 800 9000
5
10
15
20
25
30
35
40
45ab
sorb
ance
s
n
.- individual anal., +- simoult. anal., --
0 10 20 30 40 50 60 700.01
0.02
0.03
0.04
0.05
0.06
0.07
0.08
0.09
0.1
spectra channel
IR a
bsor
banc
e
IR raw process data: one run
0 10 20 30 40 50 60 70-10
-8
-6
-4
-2
0
2
4x 10
-4
spectra channel
sign
al s
econ
d de
rivat
ive
2nd derivative raw process data
0 10 20 30 40 50 60 70-10
-8
-6
-4
-2
0
2
4x 10
-4
spectra channel
sign
al s
econ
d de
rivat
ive
PCA filtered (3 comp.) 2nd derivative raw process data
0 20 40 60 80 100 120 1400
0.05
0.1
0.15
0.2
0.25
0.3
0.35
time
conc
entra
tion,
a.u
.
EFA of 2nd derivative data: initial estimation of process profiles
for 3 components
0 10 20 30 40 50 60 700
1
2
3
4
5
6
7
spectra channel
abso
rban
ce, a
.u.
1
2
3
ALS resolved pure IR spectra profiles
0 100 200 300 400 500 600 700 8000
0.05
0.1
0.15
0.2
0.25
time
conc
entra
tion,
a.u
.
1 1
1
1
11
1
3
13
1
2
2
33 2
3
2222
ALS resolved pure concetration profilesin the simultaneous analysis of eigth
runs of the process
Mixture analysis of nucleic acid constituents bymeans of continuous flow methods
Acid-base properties of the compounds analyzed
Compound pKa Working conditionscytosine 4.56 (N(3)) 37oC; I=0.15 M
11.7 (N(1)) 25oC; I=0.1 Mcytidine 4.13 (N(3)) 37oC; I=0.15 Madenosine 3.6 (N(1)) 25oC; I=0.1 Muridine 8.855 (N(3)) 37oC; I=0.15 Minosine 8.50 (N(1)) 37oC; I=0.15 Minosine-5’-monophosphate 6.00 (phosphate) 34oC; 0.5 M
9.27 (N(1)) 34oC; 0.5 M
Absorption spectra for the protonated and deprotonatedforms of the nucleosides adenosine, cytidine, inosine anduridine
200 250 300 3500
0.05
0.1
0.15
0.2
0.25
Wavelength (nm)
Abs
orba
nce
HA
HC
HI HU
200 250 300 3500
0.05
0.1
0.15
0.2
0.25
Wavelength (nm)
Abs
orba
nce
C
A
I
U
RESOLUTION OF THE EXPERIMENTAL DATA WITH
THE MCR-ALS METHOD
. Compliance of the linear Beer’s law:
Daug = Caug ST + E
Spectra
t1
tm
=
NC
NC
�n �n����
t1
t1
t1
tm
tm
tm
Sample
Standard
Csam.
Cstd.
Species correspondence between matrices
Continuous-flow system
1 1 1 1 1 1 Sample matrix
1 1 1 0 0 0 Standard matrix
Single injection flow-injection system
1 1 1 1 Sample matrix
1 1 0 0 Standard matrix
Initial estimation of the Caug data matrix
Spectra
t1
tm
NC
NC
�n �n����
t1
t1
t1
tm
tm
tm
Sample
Standard
xxxxxxxxxx
xxxxxxxxxx
xxxxx00000
xxxxx00000
ALS optimization
1st step: ST = (Caug)+ D*aug
2nd step: Caug = D*aug (ST)+
These steps are repeated until convergence is achieved.
Constraints applied:
(a) unimodal concentration profiles (not applicable to single injection
flow-injection system)
(b) non-negative concentration profiles and pure spectra
(c) the pure spectrum of each species is the same in all runs
(d) concentration profiles of common species present in different runs
have equal shapes.
RESULTS: CONTINUOUS-FLOW SYSTEM
250 300 3500
0.2
0.4
0.6
0.8
1
Wavelenght (nm)
Abs
orba
nce
interf.
interf.
interf.
HC
C
DC
0 50 1000
0.5
1
1.5
Time (seconds)
Arb
itrar
y co
ncen
tratio
n
DC
CHC
standard [C]
0 50 1000
0.5
1
1.5
Time (seconds)
Arb
itrar
y co
ncen
tratio
n
HCC DC
interf.
interf.interf.
mixture [CIAU]
Data matrix %Prediction error %fitting error
Acidic Basic Decomposition product
[CIAU;C] 9.2 51.6 7.8 1.45
[CU;C] 1.9 14.9 1.4 1.07
[CIA;C] 3.6 1.7 0.2 1.31
[drug;S] 2.4 8.6 1.1 1.74
RESULTS: SINGLE INJECTION FLOW-INJECTION
250 300 3500
0.01
0.02
0.03
0.04
0.05
0.06
0.07
0.08
Wavelength (nm)
Abs
orba
nce
interf
interf
C
HC
0 100 2000
0.5
1
1.5
2
Time (s)
Arb
itrar
y co
ncen
tratio
n
HC
C
interf
interf
mixture [CIAU]
0 100 2000
0.5
1
1.5
2
Time (s)
Arb
itrar
y co
ncen
tratio
n
C
HC
standard [C]
Data matrix Prediction error %fitting error
Acidic Basic
[CIAU;C] 1.1 1.3 2.61
[CU;C] 2.8 0.3 6.68
[CIA;C] 17.4 0.5 4.78
[drug;S] 3.4 3.7 5.31
RANK-DEFICIENT MATRICES
A data matrix is rank-deficient when the number of significant contributions to the datavariance is lower than the number of chemical components.
In closed reacting systems in which the different concentration profiles are linearlydependent due to the governing equilibrium reactions:
Rank (D) � min (R+1, S)
R = number of independent reactions; S = number of species
Acid-base processes:
System R S Rank
1 analyte 1 2 2 Full-rank
2 analytes 2 4 3 Rank-deficient
3 analytes 3 6 4 Rank-deficient
RANK-AUGMENTATION BY MATRIX-AUGMENTATION
The inclusion of the matrix of the standard in a rank-deficient system
increases the rank of the system by one:Sample Rank of the matrices
Individual Augmented
Single analyte 2 2 Full-rank
Binary mixture 3 4 Full-rank
Ternary mixture 4 5 Rank-deficient
Quaternary mixture 5 6 Rank-deficient
For quantitative purposes only partial resolution involving those species to
be quantified is required.
This partial resolution is already achieved when the analysis is performed o
the augmented matrix which contains the data matrix of the sample and tha
of the standard.
n
t
For quantitative purposes only partial resolution involving those
species to be quantified is required.
This partial resolution is already achieved when the analysis is
performed on the augmented matrix which contains the data
matrix of the sample and that of the standard.
R.Tauler, R.Gargallo, M.Vives and A.Izquierdo-Ridorsa
Chemometrics and Intelligent Lab Systems, 1998
10 20 30 40 50 60 70 80 900.55
0.6
0.65
0.7
0.75
0.8
0.85
0.9
0.95
Temperature (oC)
Abs
orba
nce
Melting data recorded at 260 nm(univariate data analysis)
20 30 40 50 60 70 80 900.23
0.235
0.24
0.245
0.25
0.255
0.26
0.265
Temperature (oC)
Abso
rban
ce
Melting recorded at 280 nm
Melting 1Melting 2
20 30 40 50 60 70 80 900
0.1
0.2
0.3
0.4
0.5
0.6
0.7
0.8
0.9
1
Temperature
Rel
ativ
e co
ncen
tratio
n
poly(A)-poly(U) ds
poly(U) rc
poly(A)-poly(U)-poly(U) tspoly(A) cs
poly(A) rc
poly(A)-poly(U) system. Two different melting experiments
ALS recovered concentrationprofiles
ALS recovered pure spectra
240 260 280 3000
0.05
0.1
0.15
0.2poly(A)-poly(U)-poly(U) ts
240 260 280 3000
0.05
0.1
0.15
0.2poly(A)-poly(U) ds
240 260 280 3000
0.05
0.1
0.15
0.2poly(U)
240 260 280 3000
0.05
0.1
0.15
0.2
poly(A)
rc
ss
Example: Two-phase induction of the non-native �-helical form of the �-
lactoglobulin in the presence of trifluoroethanol (Biophys. J.)
-40-30-20-10
0
10
20
30
40
50
0
10
20
30
40
210220
230240
250
[��]
x103
(deg
·cm
2 ·dm
ol-1
)
% TFE
/H 2O (v
/v)� (nm) nr. of components
1 2 3 4 5 6 7 8
Sing
ular
Val
ues
0
50
1001000
1100
nr. of components1 3 5 7
ln (S
.V.)
0
2
4
6
8a)b)
Mendieta J, Folque H. and Tauler R., Biophys. J., in pressMendieta J, Folque H. and Tauler R., Biophys. J., in press
��(nm)
200 210 220 230 240 250
�����x
103 (d
eg·c
m2 ·d
mol
-1)
-40-30-20-10
01020304050
I
IIIII
a)
% TFE/H2O (v/v)
0 5 10 15 20 25 30 35 40
rela
tive
conc
.
0.0
0.2
0.4
0.6
0.8
1.0IIII II
b)
Table 1: Secondary structure of the spectra recovery after ALS optimizationnative like form Intermediate form non-native helical form
�-helix 21.14 56.11 79.61�-sheet 53.81 2.14 0.00random coil 25.05 41.75 20.39
* The content of secondary structure was evaluated by least squares deconvolution using as basis spectracorresponding to �-helix, �-sheet and random coil, those described by Chen et al. (ref 36)
Mendieta J, Folque H. and Tauler R., Biophys. J., in pressMendieta J, Folque H. and Tauler R., Biophys. J., in press
% TFE/H2O (v/v)0 10 20 30 40 50 60
rela
tive
conc
.
0.0
0.2
0.4
0.6
0.8
1.0
�-sheet
random coil
�-helix
Mendieta J, Folque H. and Tauler R., Biophys. J., in pressMendieta J, Folque H. and Tauler R., Biophys. J., in press
Example: Cadmium binding properties of the C-terminal
hexapeptide from mouse methylthionein (submited for
publication)
0
2
4
6
8
10
12
14
0.0
0.5
1.01.5
2.0
-0.8-0.7
-0.6
Current (nA
)
[Cd]/[FT]Potential (V)
Number of Components0 1 2 3 4 5 6 7 8 9 10
Nor
mal
ized
Sin
gula
r Val
ues
0.0
0.1
0.2
0.8
0.9
1.0
Mendieta J, Diaz-Cruz M.S., Monjonell A., Tauler R., Esteban M., submittedMendieta J, Diaz-Cruz M.S., Monjonell A., Tauler R., Esteban M., submitted
Potential (V)-0.5 -0.6 -0.7 -0.8 -0.9 -1.0
0.00
0.25
0.50
0.75
1.00 V I IV III IIa
[Cd]/[FT]0.0 0.5 1.0 1.5 2.0
Curr
ent (
nA)
0
5
10
15
20
pH3 4 5 6 7 8 9 10 11 12
Pote
ntia
l (V)
-1.2
-1.1
-1.0
-0.9
-0.8
-0.7
-0.6
-0.5
II
III
IV
V
Mendieta J, Diaz-Cruz M.S., Monjonell A., Tauler R., Esteban M., submitted
-16
-12
-8
-4
0
4
0.00.2
0.40.60.81.01.2
200 220 240 260 280
��
[Cd]
/[FT]
wavelength�(nm)
wavelength (nm)200 220 240 260 280
��
-18
-14
-10
-6
-2
2
220 240 260 280
-2
-1
0
1
Mendieta J, Diaz-Cruz M.S., Monjonell A., Tauler R., Esteban M., submitted
0 10 20 30 40 50 60 70 80 90 1000
1
2
3
4
5
6
chemical compounds
stan
dard
ized
con
cent
ratio
ns
concentration of 96 organic compounds in 22 sampling sites of Northwestern Mediterranean Sea