CNAfingerprint

Copy number alterations (CNAs) are a predominant source of genetic alterations in human cancer and play an important role in cancer progression.

In the previous work (Wang et al. and Tao et al. ), we developed a mechanism-agnostic method to categorize CNA based on various fragment properties, which reflect the consequences of mutagenic processes and can be extracted from different types of data (WGS, SNP array), also the low-cost shallow WGS, this allowed us to get information about the tumor at a lower price!

So we developed CNAfingerprint, a tumor genomic DNA copy number alteration (CNA) based tool for assisting in clinical diagnosis and medication guidance of tumors.

As of today, CNAfingerprint can be implemented as follows:

1, target="OXA", clinical response of Oxaliplatin-based chemotherapy in metastatic colorectal cancer (mCRC).

2, target="BCG", clinical response of bacillus Calmette-Guérin (BCG) perfusion therapy in non-muscle invasive bladder cancer (NMIBC).

3. ...

we also developed a pan cancer homologous recombination deficiency (HRD) predictor, find it here!

Getting started

---

Requirements:

R (>= 3.5.0);

dplyr;

sigminer;

xgboost;

data.table (>= 1.15.0)

mlr3

carte.

devtools::install_github("XSLiuLab/CNAfingerprint")

library(CNAfingerprint)

Input format

The input requires absolute copy number profile with following information: chromosome, start, end,segVel, sample ;

The input data can be result from many software (ABSOLUTE, Sequenza, FACETS, CNVkit ...).

exampleSeg <- readRDS(system.file("extdata", "exampleSeg.rds",package = "CNAfingerprint", mustWork = TRUE))
head(exampleSeg)

#   chromosome    start       end segVal sample
# 1       chr1   900001   9500000      5 sample
# 2       chr1  9600001 248900000      6 sample
# 4      chr11   200001  18400000      7 sample
# 6      chr12   100001  41900000      6 sample
# 7      chr12 41900001  43100000      5 sample
# 8      chr12 43100001 133275309      6 sample

Extracting CNA features

features <- CNF_call(exampleSeg,hg="hg38")
head(features)[1:2,1:4]

#    sample n_of_seg n_of_cnv n_of_amp
# 1:  sample       51       51       51
# 2: sample2       52       52       51

Prediction

Once we have the CNA features and their counts, we can use it for predicting your target.

Predict the clinical response of Oxaliplatin-based chemotherapy in mCRC.

# clinical response of Oxaliplatin-based chemotherapy in mCRC

score <- CNAfingerprint(features,target="OXA")
head(score)

#  sample CNAfingerprint
# 1  sample     0.07154381
# 2 sample2     0.06783023

when CNAfingerprint_mCRC > 0.58, we consider that the patient may benefit from oxaliplatin-based chemotherapy.

---

Predict the clinical response of bacillus Calmette-Guérin (BCG) perfusion therapy in NMIBC.

# clinical response of Oxaliplatin-based chemotherapy in NMIBC

score <- CNAfingerprint(features,target="BCG")
head(score)

#    sample CNAfingerprint
# 1  sample      0.2906596
# 2 sample2      0.5066754

when CNAfingerprint > 0.38, we consider that the patient may experience recurrence after BCG treatment.

note

1,Be careful about sample ploidy selection.

2,Hg38 is the only supported version of the genome, you can convert hg19 to hg38 with the following methods:

library(GenomicRanges)
library(liftOver)

cytoarm <- cytobandToArm(ucsc.hg38.cytoband)
ch <- import.chain('hg19ToHg38.over.chain')

exampleSeg <- readRDS(system.file("extdata", "exampleSeg.rds",package = "CNAfingerprint", mustWork = TRUE))

segdata <- exampleSeg[exampleSeg$sample=="sample2",]
seg <- GRanges(seqnames = segdata$chromosome,
  ranges = IRanges(start = segdata$start,end = segdata$end),
  strand = "*",
  TCN = segdata$segVal)
  
hg38.seg <- liftOver(seg, ch)

download chain file from https://hgdownload.soe.ucsc.edu/goldenPath/hg38/liftOver/

Citation

Yao, H. et al. Copy number alteration features in pan-cancer homologous recombination deficiency prediction and biology. Commun Biol 6, 527 (2023). https://doi.org/10.1038/s42003-023-04901-3

Jy Wang. et.al. Copy number alteration fingerprint predicts the clinical response of oxaliplatin-based chemotherapy in metastatic colorectal cancer. [Under review]

Tt Cai. et.al. Copy number alteration feature predicts the clinical response of BCG perfusion therapy in non-muscle invasive bladder cancer. [Under review]

Contributors

CNAfingerprint was developed by Zy Tao, Jy Shen and Jy Wang. Please contact Jy Wang: wangjy10@shanghaitech.edu.cn for any questions or suggestions. Thank you for your use and feedback.

---

Cancer Biology Group @ShanghaiTech

Research group led by Xue-Song Liu in ShanghaiTech University