Lobry & Gautier (1994) NAR 22:3174

This page allows for the on-line reproduction of the figures in the paper: Lobry, J.R., Gautier, C. (1994) Hydrophobicity, expressivity and aromaticity are the major trends of amino-acid usage in 999 Escherichia coli chromosome-encoded genes. Nucleic Acids Research, 22:3174-3180 [PDF].

Abstract: Multivariate analysis of the amino-acid compositions of 999 chromosome-encoded proteins from Escherichia coli showed that three main factors influence the variability of amino-acid composition. The first factor was correlated with the global hydrophobicity of proteins, and it discriminated integral membrane proteins from the others. The second factor was correlated with gene expressivity, showing a bias in highly expressed genes towards amino-acids having abundant major tRNAs. Just as highly expressed genes have reduced codon diversity in protein coding sequences, so do they have a reduced diversity of amino-acid choice. This showed that translational constraints are important enough to affect the global amino-acid composition of proteins. The third factor was correlated with the aromaticity of proteins, showing that aromatic amino-acid content is highly variable.

Note: because when MV = λV we have also M(-V) = λ(-V), the signs for factors are meaningless, so that miror figures, with respect to a line thru the origin, could be obtained as well when factorial scores are involved.

ec <- read.table( file = "ftp://pbil.univ-lyon1.fr/pub/datasets/NAR94/data.txt", header = TRUE, row.names = 1) # to read the dataset ec.coa <- dudi.coa(ec, scann = FALSE, nf=3) plot(1:19, 100*ec.coa$eig/sum(ec.coa$eig), t="h", lwd=6, ylim=c(0,17), ylab="% of total variance",xlab="Factors", main="Fig 1 from Lobry & Gautier (1994) NAR 22:3174")

---

ec <- read.table( file = "ftp://pbil.univ-lyon1.fr/pub/datasets/NAR94/data.txt", header = TRUE, row.names = 1) # to read the dataset ec.coa <- dudi.coa(ec, scann = FALSE, nf=3) F1 <- ec.coa$li[,1] F1.h <- hist(F1, breaks = seq(min(F1), max(F1), le=200), plot=F) plot(F1.h$m, F1.h$c, t="h",xlab="F1",ylab="Protein count", main="Fig 2 from Lobry & Gautier (1994) NAR 22:3174")

---

ec <- read.table( file = "ftp://pbil.univ-lyon1.fr/pub/datasets/NAR94/data.txt", header = TRUE, row.names = 1) # to read the dataset ec.coa <- dudi.coa(ec, scann = FALSE, nf=3) ecfr <- ec/apply(ec, 1, sum) gravy <- read.table( file ="ftp://pbil.univ-lyon1.fr/pub/datasets/NAR94/gravy.txt") coef <- gravy$V2 grdist <- apply( ecfr[ ,order(names(ec))], 1, function(x) { sum(coef*x) }) plot(grdist, ec.coa$li[,1], pch=20, xlab="GRAVY Score", ylab="F1 Score", main="Fig 3 from Lobry & Gautier (1994) NAR 22:3174")

---

ec <- read.table( file = "ftp://pbil.univ-lyon1.fr/pub/datasets/NAR94/data.txt", header = TRUE, row.names = 1) # to read the dataset ec.coa <- dudi.coa(ec, scann = FALSE, nf=3) gravy <- read.table( file ="ftp://pbil.univ-lyon1.fr/pub/datasets/NAR94/gravy.txt") coef <- gravy$V2 x <- ec.coa$c1[order(names(ec)),1] plot(x, coef, pch = 20, xlab="First factor", ylab="GRAVY", xlim=c(min(x),max(x)+0.2), ylim=c(min(coef),max(coef)+0.2), main="Fig 4 from Lobry & Gautier (1994) NAR 22:3174") text(x+0.2,coef+0.2, lab=sort(names(ec)))

---

ec <- read.table( file = "ftp://pbil.univ-lyon1.fr/pub/datasets/NAR94/data.txt", header = TRUE, row.names = 1) # to read the dataset ec.coa <- dudi.coa(ec, scann = FALSE, nf=3) tmp <- read.table( file = "ftp://pbil.univ-lyon1.fr/pub/datasets/NAR94/ecoli999.cai") cai <- exp(tmp$V2) plot(cai, ec.coa$li[,2], pch=20, xlab="CAI Score", ylab="F2 Score", main="Fig 5 from Lobry & Gautier (1994) NAR 22:3174")

---

ec <- read.table( file = "ftp://pbil.univ-lyon1.fr/pub/datasets/NAR94/data.txt", header = TRUE, row.names = 1) # to read the dataset ec.coa <- dudi.coa(ec, scann = FALSE, nf=3) trna <- read.table( sep = "\t", file= "ftp://pbil.univ-lyon1.fr/pub/datasets/NAR94/trna.txt") x <- ec.coa$c1[order(names(ec)),2] y <- trna$V2 plot(x, y, pch=20, xlab="Second Factor", ylab="Relative amount of major tRNA (%)", xlim=c(-2.5,3.5), ylim=c(1,7), main="Fig 6 from Lobry & Gautier (1994) NAR 22:3174") text(x+0.2, y+0.2, lab= sort(names(ec)))

---

ec <- read.table( file = "ftp://pbil.univ-lyon1.fr/pub/datasets/NAR94/data.txt", header = TRUE, row.names = 1) # to read the dataset ec.coa <- dudi.coa(ec, scann = FALSE, nf=3) ecpc <- 100*ec/apply(ec, 1, sum) aro <- apply(ecpc[ ,c("phe","tyr","trp")], 1, sum) plot(aro, ec.coa$li[,3],pch=20, xlab="% aromatic amino-acid", ylab="F3 Score", main="Fig 7 from Lobry & Gautier (1994) NAR 22:3174")

---

ec <- read.table( file = "ftp://pbil.univ-lyon1.fr/pub/datasets/NAR94/data.txt", header = TRUE, row.names = 1) # to read the dataset ec.coa <- dudi.coa(ec, scann = FALSE, nf=3) tmp <- read.table( file = "ftp://pbil.univ-lyon1.fr/pub/datasets/NAR94/ecoli999.cai") cai <- exp(tmp$V2) cai.h <- hist(cai, breaks = seq(min(cai), max(cai), le=150), plot=F) plot(cai.h$m, cai.h$c, t="h",xlab="CAI Score",ylab="Protein count", xlim=c(0.1,0.9), lwd=2, main="Fig 8a from Lobry & Gautier (1994) NAR 22:3174") abline(v=quantile(cai,0.10), lty=2) abline(v=quantile(cai,0.90), lty=2)

---

ec <- read.table( file = "ftp://pbil.univ-lyon1.fr/pub/datasets/NAR94/data.txt", header = TRUE, row.names = 1) # to read the dataset tmp <- read.table( file = "ftp://pbil.univ-lyon1.fr/pub/datasets/NAR94/ecoli999.cai") cai <- exp(tmp$V2) ecfr <- ec/apply(ec, 1, sum) first <- ecfr[cai<quantile(cai,0.10),] last <- ecfr[cai>quantile(cai,0.90),] trna <- read.table( sep = "\t", file= "ftp://pbil.univ-lyon1.fr/pub/datasets/NAR94/trna.txt") coef <- trna$V2 first.d <- apply( first[ ,order(names(ec))], 1, function(x) { sum(coef*x, na.rm=T) }) last.d <- apply( last[ ,order(names(ec))], 1, function(x) { sum(coef*x, na.rm=T) }) hist( first.d, xlim=c(3.4, 5.2), border="red", density=10, col="red", xlab = "mean major tRNA frequency", ylab="Protein count", main="Fig 8b from Lobry & Gautier (1994) NAR 22:3174") hist( last.d, add=T, border="blue", density=10, angle=-45, col="blue")

---