US 2010/0145897 A1Published application

Genomic classification of malignant melanoma based on patterns of gene copy number alterations

AssigneeAbbott Laboratories

InventorsDimitri Semizarov, Xin Lu, Ke Zhang, Rick R. Lesniewski

Life of this patent

Estimating the term…

Abstract

The invention is directed to methods and kits that allow for classification of malignant melanoma cells according to genomic profiles, and methods of diagnosing, predicting clinical outcomes, and stratifying patient populations for clinical testing and treatment using the same.

Claims23

Dependent claims

A method for obtaining a database of malignant melanoma genomic subgroups, the method comprising the steps of:

(a) obtaining a plurality of m samples comprising at least one MM cell;
(b) acquiring a data set comprising copy number alteration information from at least one locus from each chromosome from each sample obtained in step (a);
(c) identifying in the data set samples contaminated by normal cells and eliminating the contaminated samples from the data set, wherein the identifying and eliminating comprises:
(1) applying a machine learning algorithm tuned to parameters that represent the differences between tumor and normal samples to the data;
(2) assigning a probability score for normal cell contamination to each sample as determined by the machine learning algorithm;
(3) eliminating data from the data set for each sample scoring 50% or greater probability of containing normal cells;
(d) estimating a number of subgroups, r, in the data set by applying an unsupervised clustering algorithm using Pearson linear dissimilarity algorithm to the data set;
(e) assigning each sample in the data set to at least one cluster using a modified genomic Non-negative Matrix Factorization (gNMF) algorithm, wherein the modified gNMF algorithm comprises:
(1) calculating divergence of the algorithm after every 100 steps of multiplicative updating using the formula:
D

(
V




WH
)
=
∑
i
=
1
n

∑
j
=
1
m

(
V
ij

log


V
ij
(
WH
)
ij
-
V
ij
+
(
WH
)
ij
)
(
11
)
wherein the Vij is the ith row and jth column of matrix V, (WH)ij is the ith row and jth column of matrix (W*H), i runs from 1 to n and n is the number of segments in the data set, and j runs from 1 to m and m is the number of samples in the data set;
(2) stopping the algorithm if the divergence calculated in step (e) (1) does not decrease by more than about 0.001% when compared to the divergence calculated for the previous 100 steps of multiplicative updating of the algorithm;
(3) randomly repeating the algorithm for a selected number of runs and calculating a Pearson correlation coefficient matrix of H for the each of run the algorithm using the formula:
C
i
,
j
=
ρ

(
H
,
i
,
H
,
j
)
=
1
r
-
1

∑
k

(
H
k
,
i
-
H
_
,
i
)

(
H
k
,
j
-
H
_
,
j
)
S
H
,
i

S
H
,
j
(
12
)
wherein C is a correlation matrix, Ci,j is an row and jth column in matrix C, H,i and H,j are an ith and jth column vector in matrix H, ρ(H,i, H,j) is a Pearson correlation coefficient between H,i and H,j, i and j run from 1 to m, m is the number of samples in the data set, k runs from 1 to r, and r is the number of subgroups from step (d);
(4) averaging the Pearson correlation coefficient matrices for each run of the algorithm obtained from step (e)(3) to arrive at an average correlation matrix;
(5) assigning tumors and cell lines in the data set into r subgroups by applying a unsupervised clustering algorithm using 1 minus the average correlation matrix determined in step (e)(4) and cutting the dendrogram into r clusters;
(f) applying a Cophenetic correlation, Bayesian Information Criterion or a combination thereof to provide a final number of clusters from the data set, wherein each final cluster defines a genomic subgroup for each tumor or cancer cell line sample;
and
(g) optionally evaluating the stability of the final number of clusters selected in step (f) using a ten-fold stability test.

123 more elements

A method of classifying a MM tumor or cell line, comprising:

(a) providing a database, developed through a method comprising:
(i) obtaining a plurality of m samples comprising at least one MM tumor or cell line;
(ii) acquiring a first data set comprising copy number alteration information from at least one locus from each chromosome from each sample obtained in step (i);
(iii) identifying in the first data set samples contaminated by normal cells and eliminating the contaminated samples from the first data set, wherein the identifying and eliminating comprises:
(1) applying a machine learning algorithm tuned to parameters that represent the differences between tumor and normal samples to the data;
(2) assigning a probability score for normal cell contamination to each sample as determined by the machine learning algorithm;
(3) eliminating data from the first data set for each sample scoring 50% or greater probability of containing normal cells;
(iv) estimating a number of subgroups, r, in the data set by applying an unsupervised clustering algorithm using Pearson linear dissimilarity algorithm to the data set;
(v) assigning each sample in the data set to at least one cluster using a modified genomic Non-negative Matrix Factorization (gNMF) algorithm, wherein the modified gNMF algorithm comprises:
(1) calculating divergence of the algorithm after every 100 steps of multiplicative updating using the formula:
D

(
V




WH
)
=
∑
i
=
1
n

∑
j
=
1
m

(
V
ij

log


V
ij
(
WH
)
ij
-
V
ij
+
(
WH
)
ij
)
(
11
)
wherein the Vij is the ith row and jth column of matrix V, (WH)ij is the ith row and jth column of matrix (W*H), i runs from 1 to n and n is the number of segments in the data set, and j runs from 1 to m and m is the number of samples in the data set;
(2) stopping the algorithm if the divergence calculated in step (v) (1) does not decrease by more than about 0.001% when compared to the divergence calculated for the previous 100 steps of multiplicative updating of the algorithm;
(3) randomly repeating the algorithm for a selected number of runs and calculating a Pearson correlation coefficient matrix of H for each of run the algorithm using the formula:
C
i
,
j
=
ρ

(
H
,
i
,
H
,
j
)
=
1
r
-
1

∑
k

(
H
k
,
i
-
H
_
,
i
)

(
H
k
,
j
-
H
_
,
j
)
S
H
,
i

S
H
,
j
(
12
)
wherein C is a correlation matrix, Ci,j is an row and jth column in matrix C, H,i and H,j are an ith and jth column vector in matrix H, ρ(H,i, H,j) is a Pearson correlation coefficient between H,i and H,j,i and j run from 1 to m, m is the number of samples in the data set, k runs from 1 to r, and r is the number of subgroups from step (iv);
(4) averaging the Pearson correlation coefficient matrices for each run of the algorithm obtained from step (v)(3) to arrive at an average correlation matrix;
(5) assigning tumors and cell lines in the data set into r subgroups by applying a unsupervised clustering algorithm using 1 minus the average correlation matrix determined in step (v)(4) and cutting the dendrogram into r clusters;
(vi) applying a Cophenetic correlation, Bayesian Information Criterion or a combination thereof to provide a final number of clusters from the data set, wherein each final cluster defines a genomic subgroup for each tumor or cancer cell line sample;
and
(vii) optionally evaluating the stability of the final number of clusters selected in step (vi) using a ten-fold stability test.
(b) providing a sample suspected of containing MM cells,
(c) acquiring a second data set, Vsample, comprising copy number alteration information from the same at least one locus from step (ii);
and
(d) classifying the sample from Vsample, by comparing Vsample to the clusters determined in steps (i)-(vii).

128 more elements

A method of classifying a therapeutic intervention for arresting or killing malignant melanoma (MM) cells, comprising:

(a) from a panel of MM cells classified according to genomic subgroups, selecting at least one MM cell line from each subgroup, wherein the panel is assembled from a method comprising:
(i) obtaining a plurality of m samples comprising MM cells;
(ii) acquiring a first data set comprising copy number alteration information from at least one locus from each chromosome from each sample obtained in step (i);
(iii) identifying in the first data set samples contaminated by normal cells and eliminating the contaminated samples from the first data set, wherein the identifying and eliminating comprises:
(1) applying a machine learning algorithm tuned to parameters that represent the differences between tumor and normal samples to the data;
(2) assigning a probability score for normal cell contamination to each sample as determined by the machine learning algorithm;
(3) eliminating data from the first data set for each sample scoring 50% or greater probability of containing normal cells;
(iv) estimating a number of subgroups, r, in the data set by applying an unsupervised clustering algorithm using Pearson linear dissimilarity algorithm to the data set;
(v) assigning each sample in the data set to at least one cluster using a modified genomic Non-negative Matrix Factorization (gNMF) algorithm, wherein the modified gNMF algorithm comprises:
(1) calculating divergence of the algorithm after every 100 steps of multiplicative updating using the formula:
D

(
V




WH
)
=
∑
i
=
1
n

∑
j
=
1
m

(
V
ij

log


V
ij
(
WH
)
ij
-
V
ij
+
(
WH
)
ij
)
(
11
)
wherein the Vij is the ith row and jth column of matrix V, (WH)ij is the ith row and jth column of matrix (W*H), i runs from 1 to n and n is the number of segments in the data set, and j runs from 1 to m and m is the number of samples in the data set;
(2) stopping the algorithm if the divergence calculated in step (v) (1) does not decrease by more than about 0.001% when compared to the divergence calculated for the previous 100 steps of multiplicative updating of the algorithm;
(3) randomly repeating the algorithm for a selected number of runs and calculating a Pearson correlation coefficient matrix of H for the each of run the algorithm using the formula:
C
i
,
j
=
ρ

(
H
,
i
,
H
,
j
)
=
1
r
-
1

∑
k

(
H
k
,
i
-
H
_
,
i
)

(
H
k
,
j
-
H
_
,
j
)
S
H
,
i

S
H
,
j
(
12
)
wherein C is a correlation matrix, Ci,j is an ith row and jth column in matrix C, H,i and H,j are an ith and jth column vector in matrix H, ρ(H,i, H,j) is a Pearson correlation coefficient between H,i and H,j, i and j run from 1 to m, m is the number of samples in the data set, k runs from 1 to r, and r is the number of subgroups from step (iv);
(4) averaging the Pearson correlation coefficient matrices for each run of the algorithm obtained from step (v)(3) to arrive at an average correlation matrix;
(5) assigning the samples in the data set into r subgroups by applying a unsupervised clustering algorithm using 1 minus the average correlation matrix determined in step (v)(4) and cutting the dendrogram into r clusters;
(vi) applying a Cophenetic correlation, Bayesian Information Criterion or a combination thereof to provide a final number of clusters from the data set, wherein each final cluster defines a genomic subgroup for each sample;
and
(vii) optionally evaluating the stability of the final number of clusters selected in step (vi) using a ten-fold stability test;
(viii) selecting at least one MM cell from each cluster selected in step (vi) and assembling into panels defined according to genomic subgroups.
(b) contacting the at least one MM cell from each subgroup with the therapeutic intervention;
(c) assaying the effectiveness of the therapeutic intervention to arrest or kill the at least one MM cell from each subgroup;
(d) classifying the therapeutic intervention according to the determined effectiveness of the therapeutic intervention to arrest or kill the at least one MM cell from each subgroup,
wherein arresting or killing the at least one MM cell from one subgroup, but not another indicates specificity of the therapeutic intervention to arrest or kill MM cells of that subgroup.

129 more elements

A method of assembling a probe panel for classifying a MM cell from a sample, comprising:

(a) assembling a database, comprising:
(i) obtaining a plurality of m samples comprising at least one MM cell;
(ii) acquiring a first data set comprising copy number alteration information from at least one locus from each chromosome from each sample obtained in step (i);
(iii) identifying in the first data set samples contaminated by normal cells and eliminating the contaminated samples from the first data set, wherein the identifying and eliminating comprises:
(1) applying a machine learning algorithm tuned to parameters that represent the differences between tumor and normal samples to the data;
(2) assigning a probability score for normal cell contamination to each sample as determined by the machine learning algorithm;
(3) eliminating data from the first data set for each sample scoring 50% or greater probability of containing normal cells;
(iv) estimating a number of subgroups, r, in the data set by applying an unsupervised clustering algorithm using Pearson linear dissimilarity algorithm to the data set;
(v) assigning each sample in the data set to at least one cluster using a modified genomic Non-negative Matrix Factorization (gNMF) algorithm, wherein the modified gNMF algorithm comprises:
(1) calculating divergence of the algorithm after every 100 steps of multiplicative updating using the formula:
D

(
V




WH
)
=
∑
i
=
1
n

∑
j
=
1
m

(
V
ij

log


V
ij
(
WH
)
ij
-
V
ij
+
(
WH
)
ij
)
(
11
)
wherein the Vij is the ith row and jth column of matrix V, (WH)ij is the ith row and jth column of matrix (W*H), i runs from 1 to n and n is the number of segments in the data set, and j runs from 1 to m and m is the number of samples in the data set;
(2) stopping the algorithm if the divergence calculated in step (v) (1) does not decrease by more than about 0.001% when compared to the divergence calculated for the previous 100 steps of multiplicative updating of the algorithm;
(3) randomly repeating the algorithm for a selected number of runs and calculating a Pearson correlation coefficient matrix of H for the each of run the algorithm using the formula:
C
i
,
j
=
ρ

(
H
,
i
,
H
,
j
)
=
1
r
-
1

∑
k

(
H
k
,
i
-
H
_
,
i
)

(
H
k
,
j
-
H
_
,
j
)
S
H
,
i

S
H
,
j
(
12
)
wherein C is a correlation matrix, Ci,j is an row and jth column in matrix C, H,i and H,j are an ith and jth column vector in matrix H, ρ(H,i, H,j) is a Pearson correlation coefficient between H,i and H,j,i and j run from 1 to m, m is the number of samples in the data set, k runs from 1 to r, and r is the number of subgroups from step (iv);
(4) averaging the Pearson correlation coefficient matrices for each run of the algorithm obtained from step (v)(3) to arrive at an average correlation matrix;
(5) assigning tumors and cell lines in the data set into r subgroups by applying a unsupervised clustering algorithm using 1 minus the average correlation matrix determined in step (v)(4) and cutting the dendrogram into r clusters;
(vi) applying a Cophenetic correlation, Bayesian Information Criterion or a combination thereof to provide a final number of clusters from the data set, wherein each final cluster defines a genomic subgroup for each tumor or cancer cell line sample;
and
(vii) optionally evaluating the stability of the final number of clusters selected in step (vi) using a ten-fold stability test.
(viii) selecting at least one sample from each cluster selected in step (vi) and assembling into panels defined according to genomic subgroups;
(b) analyzing the database of step (a) to determine characteristic copy number abnormalities for each subgroup;
(c) designing a plurality of probes based on the determined characteristic copy number abnormalities for each subgroup and assigning each probe to a genomic subgroup.

127 more elements

A kit for classifying a MM tumor sample or a cell line, comprising:

(a) instructions to assemble a database, comprising instructions for:
(i) obtaining a plurality of m samples comprising at least one MM cell;
(ii) acquiring a first data set comprising copy number alteration information from at least one locus from each chromosome from each sample obtained in step (i);
(iii) identifying in the first data set samples contaminated by normal cells and eliminating the contaminated samples from the first data set, wherein the identifying and eliminating comprises:
(1) applying a machine learning algorithm tuned to parameters that represent the differences between tumor and normal samples to the data;
(2) assigning a probability score for normal cell contamination to each sample as determined by the machine learning algorithm;
(3) eliminating data from the first data set for each sample scoring 50% or greater probability of containing normal cells;
(iv) estimating a number of subgroups, r, in the data set by applying an unsupervised clustering algorithm using Pearson linear dissimilarity algorithm to the data set;
(v) assigning each sample in the data set to at least one cluster using a modified genomic Non-negative Matrix Factorization (gNMF) algorithm, wherein the modified gNMF algorithm comprises:
(1) calculating divergence of the algorithm after every 100 steps of multiplicative updating using the formula:
D

(
V




WH
)
=
∑
i
=
1
n

∑
j
=
1
m

(
V
ij

log


V
ij
(
WH
)
ij
-
V
ij
+
(
WH
)
ij
)
(
11
)
wherein the Vij is the ith row and jth column of matrix V, (WH)ij is the ith row and jth column of matrix (W*H), i runs from 1 to n and n is the number of segments in the data set, and j runs from 1 to m and m is the number of samples in the data set;
(2) stopping the algorithm if the divergence calculated in step (v) (1) does not decrease by more than about 0.001% when compared to the divergence calculated for the previous 100 steps of multiplicative updating of the algorithm;
(3) randomly repeating the algorithm for a selected number of runs and calculating a Pearson correlation coefficient matrix of H for the each of run the algorithm using the formula:
C
i
,
j
=
ρ

(
H
,
i
,
H
,
j
)
=
1
r
-
1

∑
k

(
H
k
,
i
-
H
_
,
i
)

(
H
k
,
j
-
H
_
,
j
)
S
H
,
i

S
H
,
j
(
12
)
wherein C is a correlation matrix, Ci,j is an ith row and jth column in matrix C, H,i and H,j are an ith and jth column vector in matrix H, ρ(H,i, H,j) is a Pearson correlation coefficient between H,i and H,j, i and j run from 1 to m, m is the number of samples in the data set, k runs from 1 to r, and r is the number of subgroups from step (iv);
(4) averaging the Pearson correlation coefficient matrices for each run of the algorithm obtained from step (v)(3) to arrive at an average correlation matrix;
(5) assigning tumors and cell lines in the data set into r subgroups by applying a unsupervised clustering algorithm using 1 minus the average correlation matrix determined in step (v)(4) and cutting the dendrogram into r clusters;
(vi) applying a Cophenetic correlation, Bayesian Information Criterion or a combination thereof to provide a final number of clusters from the data set, wherein each final cluster defines a genomic subgroup for each tumor or cancer cell line sample;
and
(vii) optionally evaluating the stability of the final number of clusters selected in step;
and (vi) using an ten-fold stability test;
and
(b) optionally, a first, second, third, fourth, fifth and sixth cell lines, or isolated genomic DNA thereof, wherein
the first cell line is selected from the group consisting of SKMEL119, HS944, WM1366, and WM88;
the second cell line is WM3248;
the third cell line is 1205LU;
the fourth cell line is selected from the group consisting of 451LU, SKMEL19, SKMEL28, SKMEL30, SKMEL63, WM35, WM983, and WM983C;
the fifth cell line is selected from the group consisting of WM3211, M14, MEWO, SKMEL2, SKMEL5, UACC257, UACC62, WM122, WM13662, WM239A, WM32112, WM32482, WM793B, and 501MEL, and
the sixth cell line is MALME3M or WM882.

133 more elements

Dependencies are inferred from the claim text.

Facts

Application: 12/607,080Filed 28 Oct 2009
Priority: 31 Oct 2008
Published: 10 Jun 2010
Term ends: Not estimatedThe source records don’t support an estimate.

Classification16 codes

G16H 50/20
ICT specially adapted for medical diagnosis, medical simulation or medical data mining; ICT specially adapted for detecting, monitoring or…: for…
C12Q 1/6886
For diseases caused by alterations of genetic material: for cancer
G16B 20/10
ICT specially adapted for functional genomics or proteomics, e.g. genotype-phenotype associations: ploidy or copy number detection
G16B 20/20
ICT specially adapted for functional genomics or proteomics, e.g. genotype-phenotype associations: allele or variant detection, e.g. single…

Family

Application61/110,308Us provisional application · filed 31 Oct 2008
Application61/110,281Us provisional application · filed 31 Oct 2008
US 8,498,821 B2Related patent
US 2010/0145897 A1This patent

From the US filing history. Foreign family members aren’t listed.

Timeline

Citations

Open elsewhere

patentgrep

Loading patent…

Abstract

A method for obtaining a database of malignant melanoma genomic subgroups, the method comprising the steps of:

(a) obtaining a plurality of m samples comprising at least one MM cell;
(b) acquiring a data set comprising copy number alteration information from at least one locus from each chromosome from each sample obtained in step (a);
(c) identifying in the data set samples contaminated by normal cells and eliminating the contaminated samples from the data set, wherein the identifying and eliminating comprises:
(1) applying a machine learning algorithm tuned to parameters that represent the differences between tumor and normal samples to the data;
(2) assigning a probability score for normal cell contamination to each sample as determined by the machine learning algorithm;
(3) eliminating data from the data set for each sample scoring 50% or greater probability of containing normal cells;
(d) estimating a number of subgroups, r, in the data set by applying an unsupervised clustering algorithm using Pearson linear dissimilarity algorithm to the data set;
(e) assigning each sample in the data set to at least one cluster using a modified genomic Non-negative Matrix Factorization (gNMF) algorithm, wherein the modified gNMF algorithm comprises:
(1) calculating divergence of the algorithm after every 100 steps of multiplicative updating using the formula:
D

(
V




WH
)
=
∑
i
=
1
n

∑
j
=
1
m

(
V
ij

log


V
ij
(
WH
)
ij
-
V
ij
+
(
WH
)
ij
)
(
11
)
wherein the Vij is the ith row and jth column of matrix V, (WH)ij is the ith row and jth column of matrix (W*H), i runs from 1 to n and n is the number of segments in the data set, and j runs from 1 to m and m is the number of samples in the data set;
(2) stopping the algorithm if the divergence calculated in step (e) (1) does not decrease by more than about 0.001% when compared to the divergence calculated for the previous 100 steps of multiplicative updating of the algorithm;
(3) randomly repeating the algorithm for a selected number of runs and calculating a Pearson correlation coefficient matrix of H for the each of run the algorithm using the formula:
C
i
,
j
=
ρ

(
H
,
i
,
H
,
j
)
=
1
r
-
1

∑
k

(
H
k
,
i
-
H
_
,
i
)

(
H
k
,
j
-
H
_
,
j
)
S
H
,
i

S
H
,
j
(
12
)
wherein C is a correlation matrix, Ci,j is an row and jth column in matrix C, H,i and H,j are an ith and jth column vector in matrix H, ρ(H,i, H,j) is a Pearson correlation coefficient between H,i and H,j, i and j run from 1 to m, m is the number of samples in the data set, k runs from 1 to r, and r is the number of subgroups from step (d);
(4) averaging the Pearson correlation coefficient matrices for each run of the algorithm obtained from step (e)(3) to arrive at an average correlation matrix;
(5) assigning tumors and cell lines in the data set into r subgroups by applying a unsupervised clustering algorithm using 1 minus the average correlation matrix determined in step (e)(4) and cutting the dendrogram into r clusters;
(f) applying a Cophenetic correlation, Bayesian Information Criterion or a combination thereof to provide a final number of clusters from the data set, wherein each final cluster defines a genomic subgroup for each tumor or cancer cell line sample;
and
(g) optionally evaluating the stability of the final number of clusters selected in step (f) using a ten-fold stability test.

123 more elements

Facts

Application

12/607,080Filed 28 Oct 2009

Priority

31 Oct 2008

Published

10 Jun 2010

Term ends

Not estimatedThe source records don’t support an estimate.

Classification16 codes

G16H 50/20
ICT specially adapted for medical diagnosis, medical simulation or medical data mining; ICT specially adapted for detecting, monitoring or…: for…
C12Q 1/6886
For diseases caused by alterations of genetic material: for cancer
G16B 20/10
ICT specially adapted for functional genomics or proteomics, e.g. genotype-phenotype associations: ploidy or copy number detection
G16B 20/20
ICT specially adapted for functional genomics or proteomics, e.g. genotype-phenotype associations: allele or variant detection, e.g. single…