#
# read Sib-pair output and create a pretty HTML document that
# summarises it with links to detailed tables
#
AWK=awk
if ! [ -f $1 ] || [ "$1" = "" ]
then
echo "sp2html: summarise Sib-pair output file as a HTML document."
exit
fi
#
# Summary
#
$AWK 'BEGIN { comment="[!#]"
first_comment=0 # flag whether writing out contents of first comment
prodrome=1 # reading preliminary data
nfreq=0 # number of allele frequency tabulations
nanal=6 # number of possible analyses
analname[1]="APM"
analname[2]="GPM"
analname[3]="H-E"
analname[4]="TDT"
analname[5]="Ass"
analname[6]="ASP"
header="|||| SIB-PAIR: A program for simple genetic analysis"
}
NR==1 && $0!=header {
print "This does not seem to be output from Sib-pair!"
print "Exiting."
exit
}
#
# Initialize the HTML output file
#
NR==1 { getline
version=$NF
getline
getline
daterun=substr($0,16)
print ""
print ""
print "Summary of Sib-pair output "" \
FILENAME """
print ""
print ""
print " "
print "Summary and full text versions of Sib-pair (v" \
version ") output file "" \
FILENAME ""
"
}
#
# Copy the text of the first comment
#
first_comment<2 {
if (substr($1,1,1) ~ comment) {
if (first_comment==0 && length($0>2)) {
first_comment=1
printf "\nText of first comment
\n\n\n"
}
print $0
}else{if (substr($1,1,1) !~ comment && first_comment==1) {
first_comment=2
print ""
}}
}
$1=="Pedigree" && $2=="file" {
pedfile=$NF
}
substr($0,1,25)=="Total number of pedigrees" {
numped=$NF
}
substr($0,1,24)=="Total number of subjects" && prodrome {
prodrome=0
numrec=$NF
print " "
print ""
print " "
print ""
print "| Date run | " daterun
print " |
|---|
| Pedigree file | "" pedfile """
print " |
|---|
| No. of Individuals | " numrec
print " |
|---|
| No. of Pedigrees | " numped
print " |
|---|
"
print " "
print ""
print " "
print ""
print "
"
print "| Locus name | Locus Type"
ntrait=0
nmark=0
for (i=1;i<=numloc;i++) {
if (loctyp[i]=="m") {
nmark++
markername[nmark]=locnam[i]
typ="Marker (codominant)"
}else{if (loctyp[i]=="a") {
ntrait++
traitname[ntrait]=locnam[i]
typ="Binary trait"
}else{if (loctyp[i]=="q") {
ntrait++
traitname[ntrait]=locnam[i]
typ="Quantitative trait"
}else{
typ=" "
}}}
print " |
|---|
| " locnam[i] " | " typ
}
print " |
"
}
#
# Tabulate the loci
#
$0=="Locus Type Position" {
getline
getline
numloc=0
while (NF>0) {
numloc++
locnam[numloc]=$1
loctyp[numloc]=$2
getline
}
}
#
# Interpret the results
#
$1=="Allele" && $2=="frequencies" {
typ="Fre"
gsub(/\"/," ");
nfreq++
frename[nfreq]=$NF
getline
getline
getline
minall[nfreq]=$1
while (NF>0) {
maxall[nfreq]=$1
getline
}
getline
numalleles[nfreq]=$NF
getline
het[nfreq]=$NF
getline
getline
numtyped[nfreq]=$5
}
$1=="Hardy-Weinberg" && $2=="equilibrium" {
typ="HWE"
gsub(/\"/," ");
mrker=$NF
}
$1=="APM" {
has_res[1]=1
typ="APM"; gsub(/\"/," "); trait=$4; mrker=$8 $9; aff=0; gpm=0
}
$1=="Aff-Aff" && typ=="APM" {aff=1}
$1=="Aff-UnA" && typ=="APM" {aff=0}
$1=="GPM" && typ=="APM" { has_res[2]=1; gpm=1}
$1=="H-E" { gsub(/\"/," "); has_res[3]=1; typ="H-E"; trait=$4; mrker=$7 }
$1=="TDT" { gsub(/\"/," "); has_res[4]=1; typ="TDT"; trait=$4; mrker=$7 }
$1=="Assoc" { gsub(/\"/," "); has_res[5]=1; typ="Ass"; trait=$4; mrker=$7 }
$2=="ASP" { gsub(/\"/," "); has_res[6]=1; typ="ASP"; trait=$4; mrker=$7 }
$1=="P-value=" {
if($2<0.001){flag="+++"
}else{if($2<0.01){flag="++"
}else{if($2<0.05){flag="+"
}else{flag="0"
}}}
pval=$2
getline
if($1=="Empiric"){
empval=$NF
if(empval<0.001){flag="+++"
}else{if(empval<0.01){flag="++"
}else{if(empval<0.05){flag="+"
}}}
}
result[trait,typ,mrker]=flag
}
typ=="Ass" && $1=="Equalled" {
empval=$NF
gsub("[()]","",empval)
if(empval<0.001){flag="+++"
}else{if(empval<0.01){flag="++"
}else{if(empval<0.05){flag="+"
}else{flag="0"
}}}
result[trait,typ,mrker]=flag
}
typ=="Ass" && $2=="P-value" {
if($4<0.001){flag="+++"
}else{if($4<0.01){flag="++"
}else{if($4<0.05){flag="+"
}else{flag="0"
}}}
pval=$4
getline
if($1=="Equalled"){
empval=$NF
gsub("[()]","",empval)
if(empval<0.001){flag="+++"
}else{if(empval<0.01){flag="++"
}else{if(empval<0.05){flag="+"
}}}
}
result[trait,typ,mrker]=flag
}
typ=="H-E" && $1=="t" && $2=="value" {
pval=$NF
gsub("[()]","",pval)
gsub("P=","",pval)
if(pval<0.001){flag="+++"
}else{if(pval<0.01){flag="++"
}else{if(pval<0.05){flag="+"
}else{flag="0"
}}}
result[trait,typ,mrker]=flag
}
typ=="APM" && $1=="f(p)" && $3=="1/sqrt(p)" && aff {
pval=$6
if(pval<0.001){flag="+++"
}else{if(pval<0.01){flag="++"
}else{if(pval<0.05){flag="+"
}else{flag="0"
}}}
result[trait,typ,mrker]=flag
}
typ=="APM" && $1=="f(p)" && $3=="1/sqrt(p)" && gpm {
pval=$6
if(pval<0.001){flag="+++"
}else{if(pval<0.01){flag="++"
}else{if(pval<0.05){flag="+"
}else{flag="0"
}}}
result[trait,"GPM",mrker]=flag
}
typ=="ASP" && $1=="Full-sibs" {
if ($6>$7) { fchar="+" }else{ fchar="-" }
}
typ=="ASP" && $1=="Full-Sib" {
pval=0+substr($NF,length($NF)-5,5)
if(pval<0.001){flag=" " fchar fchar fchar
}else{if(pval<0.01){flag=" " fchar fchar
}else{if(pval<0.05){flag=" " fchar
}else{flag="0"
}}}
result[trait,typ,mrker]=flag
}
typ=="HWE" && $2=="P-value" && $1=="Nominal" {
pval=$4
if(pval<0.001){flag="+++"
}else{if(pval<0.01){flag="++"
}else{if(pval<0.05){flag="+"
}else{flag="0"
}}}
hwe_pvalue[mrker]=flag
}
typ=="HWE" && $1=="Equalled" {
empval=$NF
gsub("[()]","",empval)
if(empval<0.001){flag="+++"
}else{if(empval<0.01){flag="++"
}else{if(empval<0.05){flag="+"
}else{flag="0"
}}}
hwe_pvalue[mrker]=flag
}
#
# Write out the results
#
END { print " "
print "Results
"
printf "\n\n" \
"Coding for statistical signifance\n\n" \
"| Code | Meaning\n" \
" |
|---|
| +++" \
" | P < 0.001\n" \
" |
| ++" \
" | 0.001 < P < 0.01\n" \
" |
| +" \
" | 0.01 < P < 0.05\n" \
" |
| 0" \
" | 0.05 < P\n" \
" |
| ." \
" | Test missing\n" \
" |
\n\n"
#
# First, allele frequencies
#
if (nfreq>0) {
print ""
printf "\n
\n| Marker" \
" | Persons typed | No. of alleles" \
" | Allele (size) range | Heterozygosity | HWE test"
for(k=1;k<=nfreq;k++) {
ch=hwe_pvalue[frename[k]]
if (ch=="") {ch="."}
print " |
|---|
| " frename[k] "" \
" | " numtyped[k] \
" | " numalleles[k] \
" | " minall[k] "--" maxall[k] \
" | " het[k] \
" | " ch
}
printf " |
|---|
\n\n"
}
#
# Then trait analyses
#
print "Trait analyses
"
print " "
for(i=1;i<=ntrait;i++) {
printf "\n
\n" \
"Trait: " \
""" traitname[i] ""\n" \
"| Marker" \
" | Test\n |
|---|
"
for(k=1;k<=nanal;k++) {
if (has_res[k]) {
print "| " \
analname[k] ""
}else{
print " | " analname[k]
}
}
for(j=1;j<=nmark;j++) {
printf " |
| " markername[j]
for(k=1;k<=nanal;k++) {
ch=result[traitname[i],analname[k],markername[j]]
if (ch=="") {
ch="."
}
printf " | " ch
}
printf "\n"
}
print " |
|---|
"
}
}' $1
#
# Second pass to put in the detail
#
$AWK 'BEGIN { comment="[!#]"
in_comment=0 # within a comment
in_table=0 # within any table
freq_table=0 # printing table of allele frequencies
apm_table=0 # printing table of APM sharing
ibs_table=0 # printing table of ibs ASP sharing
tdt_table=0 # printing table of TDT results
first_freq=1 # start of set of tables of allele freqs
first_apm=1 # start of set of tables of APM results
first_ibs=1 # start of set of tables of ASP results
first_tdt=1 # start of set of tables of TDT results
last_unfo=0 # last line of unformatted output
header="|||| SIB-PAIR: A program for simple genetic analysis"
typloc["a"]="aff"
typloc["q"]="qua"
typloc["m"]="mar"
printf "\n\n\n\n"
}
#
# Remove reserved symbols
#
{ gsub("&","\\\&",$0)
gsub("<","\\\<",$0)
gsub(">","\\\>",$0)
gsub("\"","\\\"",$0)
#
# Tidy up completed blocks
#
if (in_table && $0 !~ /=/) {
in_table=0
printf "\n\n"
}else{if (in_comment && substr($1,1,1) !~ comment) {
in_comment=0
print ""
}}
#
# Perform action appropriate to current line
#
if ($0==header) {
print "
"
print "" substr($0,1,4) "" substr($0,5) "
"
getline
print "" substr($0,1,4) "" substr($0,5) "
"
getline
print "" substr($0,1,4) "" substr($0,5) "
"
getline
print "" substr($0,1,4) "" substr($0,5) "
"
print "
"
getline
#
# A comment
#
}else{if (substr($1,1,1) ~ comment) {
if (!in_comment) {
in_comment=1
print ""
print "
"
}
print $0
#
# A warning or informative message
#
}else{if ($0 ~ /ERROR:/) {
printf "\nERROR: " substr($0,6) "\n"
getline
while (length($0)>0 && $1!="Pedigree:") {
print $0
getline
}
print "\n
\n"
}else{if ($0 ~ /NOTE:/) {
printf "
\nNOTE: " substr($0,6) "\n"
getline
while (length($0)>0 && $1!="Pedigree:" && $0 !~/NOTE/) {
print $0
getline
}
print "\n
\n"
#
# The title of the current analysis
#
}else{if ($0 ~ /-------------------------------/) {
getline
freq_table=($1=="Allele" && $2=="frequencies")
ibs_table=($1=="IBS" && $2=="ASP")
tdt_table=($1=="TDT")
if (freq_table && first_freq) {
first_freq=0
printf "
"
}else{if (apm_table && first_apm) {
first_apm=0
printf ""
}else{if (tdt_table && first_tdt) {
first_tdt=0
printf ""
}else{if (ibs_table && first_ibs) {
first_ibs=0
printf ""
}else{
print "" $0 "
"
}}}}
getline
#
# List of dyads variable = value. Formatted as a table.
#
}else{if ($0 ~ /=/) {
if (!in_table) {
in_table=1
printf "\n
\n"
}
pos=index($0,"=")
printf "| " substr($0,1,pos-1) \
" | " substr($0,pos) "\n"
#
# List of loci
#
}else{if ($0=="Locus Type Position") {
printf "\n" \
" \n \n" \
"| Locus | Type | Position\n"
getline
getline
while(NF>2) {
print " |
|---|
| " $1 " | " typloc[$2] \
" | " $3 $4
getline
}
printf " | \n\n"
#
# table of allele-by-allele TDT results
#
}else{if ($0 ~ /Allele by Allele TDT/) {
getline
getline
getline
printf "\n \n\n" \
"Allele by Allele TDT\n" \
"| Allele | Trans | Not Tr | TDT | P-value\n"
while (NF==5) {
print " |
|---|
| " $1 \
" | " $2 \
" | " $3 \
" | " $4 \
" | " $5
getline
}
printf " | \n\n"
#
# table of genotypic TDT results
#
}else{if ($0 ~ /Marginal Genotypic TDT/) {
getline
getline
getline
printf "\n \n\n" \
"Marginal Genotypic TDT\n\n" \
"| Genotype | Trans | Not Tr\n"
while (NF==3) {
print " |
|---|
| " $1 \
" | " $2 \
" | " $3
getline
}
printf " | \n\n"
#
# table of global allelic TDT results
#
}else{if ($0 ~ /Global Allelic TDT/) {
getline
getline
getline
printf "\n \n\n" \
"Global Allelic TDT\n" \
"| All 1 | All 2 | Tr=1 | Tr=2 | TDT\n"
while ($1>0) {
print " |
|---|
| " $1 \
" | " $2 \
" | " $3 \
" | " $4 \
" | " $5
getline
}
printf " | \n\n"
#
# table of ibs ASP sharing
#
}else{if (ibs_table) {
printf "\n \n| " \
" | No. of Pairs" \
" | IBS Sharing" \
" | Mean IBS sharing\n" \
" |
|---|
| 2/2 | 1/2 | 0/2 | Obs | Exp\n"
getline
getline
getline
while(NF==7) {
print " |
|---|
| " $1 \
" | " $2 \
" | " $3 \
" | " $4 \
" | " $5 \
" | " $6 \
" | " $7 "\n"
getline
}
printf " | \n\n"
ibs_table=0
#
# table of allele frequencies
#
}else{if (freq_table) {
printf "\n \n" \
"| Allele | Frequency | Count | Histogram"
getline
while(NF>0) {
print " |
|---|
| " $1 \
" | " $2 \
" | " $3 \
" | " $4
getline
}
printf " | \n\n"
freq_table=0
#
# Detailed HWE table
#
}else{if ($2=="Observed" && $3=="Genotypes") {
printf "\n \n\n" \
"Observed genotype counts\n" \
"\n"\
"| Genotype | Observed | Expected | Deviate\n"
getline
getline
getline
while (NF>1) {
print " |
|---|
| " $1 \
" | " $2 \
" | " $3 \
" | " $4
getline
}
getline
printf " | | Total" \
" | " $2 "\n | \n\n"
#
# tables of segregation ratios and recurrence risks
#
}else{if ($0=="Mating Type UxU UxA AxA" || \
$0=="Relative pair RecRisk Aff-Aff Aff-UnA" || \
$0=="Total sample All Fndrs Nonfndrs") {
printf "\n \n| " $1 " " $2 \
" | " $3 " | " $4 " | " $5 "\n"
getline
getline
while (length($0)>0) {
print " |
|---|
| " substr($0,1,11) \
" | " substr($0,12,9) \
" | " substr($0,21,9) \
" | " substr($0,30,9)
getline
}
printf " | \n\n"
#
# Marker Homozygosity test
#
}else{if ($1=="Marker" && $2=="N") {
printf "\n \n" \
"\n" \
"| Marker | N" \
" | Obs | Exp | F | Z | Emp P | Iter\n"
getline
getline
while (NF>1) {
print " |
|---|
| " $1 \
" | " $2 \
" | " $3 \
" | " $4 \
" | " $5 \
" | " $6 \
" | " $7 \
" | " $8
getline
}
printf " | \n\n"
#
# Kaplan-Meier table
#
}else{if ($1=="Age-at-onset" && $2=="Failed") {
printf "\n \n" \
"| Age-at-onset | Failed" \
" | Riskset | H(t) | S(t) | ase"
getline
getline
while(NF==6) {
print " |
|---|
| " $1 \
" | " $2 \
" | " $3 \
" | " $4 \
" | " $5 \
" | " $6
getline
}
printf " | \n\n"
#
# Quantitative trait descriptive stats
#
}else{if ($1=="Descriptive" && $2=="Stats") {
printf "\n \n\n" \
"Quantitative trait descriptive statistics\n" \
"\n| Statistic" \
" | All | Founders | Nonfounders\n"
getline
getline
while (NF>0) {
print " |
|---|
| " substr($0,1,12) \
" | " substr($0,15,13) \
" | " substr($0,28,13) \
" | " substr($0,41,13)
getline
}
printf " |
|---|
\n\n"
}else{if ($2=="Familial" && $3=="correlations") {
printf "\n \n\n" \
"Familial correlations (pairwise)\n" \
"\n| Rel 1 | Rel 2" \
" | Variance 1 | Variance 2 | Correlation" \
" | N Pairs\n"
getline
getline
getline
while (NF>0) {
print " |
|---|
| " $1 " | " $2 \
" | " substr($0,15,13) \
" | " substr($0,28,13) \
" | " substr($0,41,13) \
" | " $NF
getline
}
printf " |
|---|
\n\n"
}else{if ($1=="Fain" && $2=="sibship") {
printf "\n \n\n" \
"Fain sibship variance test\n\n"
getline
getline
while (length($0)>0) {
print "| " substr($0,1,14) \
" | " substr($0,15)
getline
}
printf " | \n\n"
#
# List of pedigrees
#
}else{if ($1=="Pedigree:" && $3=="No.") {
printf "\n \n| Pedigree | No. members" \
" | No. founders | No. sibships\n"
while ($1=="Pedigree:") {
print " |
|---|
| " $2 " | " $5 " | " $8 " | " $11
getline
}
printf " | \n\n"
#
# QTL Anova
#
}else{if ($2=="QTL" && $3=="Association") {
printf "\n \n\n" \
"QTL Association Analysis\n" \
"\n| Allele" \
" | Allelic Mean | Stand Error | Count\n"
getline
getline
getline
while (NF==4) {
print " |
|---|
| " $1 \
" | " $2 \
" | " $3 \
" | " $4
getline
}
printf " | \n\n"
}else{if ($2=="Association" && $3=="Analysis") {
printf "\n \n\n" \
"Binary Trait Association Analysis\n" \
"\n| Allele" \
" | Affected | Unaffected | Total | Dev\n"
getline
getline
getline
while (NF>1) {
print " |
|---|
| " $1 \
" | " $2 " " $3 \
" | " $4 " " $5 \
" | " $6 \
" | " $7
getline
}
getline
print " | | Total" \
" | " $2 \
" | " $3 \
" | " $4
printf " | \n\n"
}else{if ($1=="->") {
$1=""
printf "\n" \
"[CMD: " $0 " ]\n" \
" \n"
#
# A pedigree drawing either single locus -- of errors, or multilocus -- of
# haplotypes
#
}else{if ($1=="Sibship:") {
print ""
print $0
print " "
getline
while (NF!=2) {
if (NF>0) {
print "" $0 "
"
}else{
print " "
}
getline
}
print " "
while (NF>0) {
print $0
getline
}
print ""
}else{if ($1=="This" && $2=="job" && $3=="took") {
printf "
\n" $0 "\n
\n"
}else{if (length($0)>0) {
if (last_unfo!=(NR-1)) {
print ""
}
last_unfo=NR
print $0 "
"
}}}}}}}}}}}}}}}}}}}}}}}}}}
}
END { printf "\n\n"
}' $1
|