def __init__(
self,
path,
lastcol=None,
lastrow=None,
colmeta=None,
rowmeta=None,
eps_noise=False,
):
self.tbl = table(path)
self.bak = self.tbl.copy()
# strip / save metadata
if lastrow is not None:
self.tbl.head(lastrow, invert=True)
if lastcol is not None:
self.tbl.head(lastcol, invert=True, transposed=True)
self.row = self.tbl.rowheads[:]
self.col = self.tbl.colheads[:]
if eps_noise:
self.tbl.float()
self.tbl.apply_entries(lambda x: x + c_eps * random.random())
self.dat = self.tbl.table2array()
# colmetas from file / table
if colmeta is None:
self.colmeta = None
self.colmetaname = None
else:
self.colmeta = []
self.colmetaname = []
for x in colmeta:
if os.path.exists(x):
warn("Loading col metadata from file:", x)
temp = col2dict(x, value=1)
self.colmeta.append(
[temp.get(k, c_str_none) for k in self.col])
self.colmetaname.append(path2name(x))
else:
temp = self.bak.rowdict(x)
self.colmeta.append(
[temp.get(k, c_str_none) for k in self.col])
self.colmetaname.append(x)
# rowmetas from file / table
if rowmeta is None:
self.rowmeta = None
self.rowmetaname = None
else:
self.rowmeta = []
self.rowmetaname = []
for x in rowmeta:
if os.path.exists(x):
warn("Loading row metadata from file:", x)
temp = col2dict(x, value=1)
self.rowmeta.append(
[temp.get(k, c_str_none) for k in self.row])
self.rowmetaname.append(path2name(x))
else:
temp = self.bak.coldict(x)
self.rowmeta.append(
[temp.get(k, c_str_none) for k in self.row])
self.rowmetaname.append(x)
def main( ):
args = get_args( )
tbl = table( args.table )
data = tbl.table2array( args.last_metadata ).transpose( )
dist = distance_matrix( data, args.dissimilarity )
if args.method == "manual":
embedding, varexp, goodness = ordinate_cmdscale( dist )
else:
embedding, varexp, goodness = ordinate_sklearn( dist, method=args.method )
fig = plt.figure()
fig.set_size_inches( 10, 6 )
ax = plt.subplot( 111 )
xcoords = embedding[:, 0]
ycoords = embedding[:, 1]
shapes = ["o" for x in xcoords]
if args.shapeby is not None:
field, path = args.shapeby
shapes = shapeize( tbl.row( field ), path, ax )
colors = ["black" for x in xcoords]
if args.colorby is not None:
field, path = args.colorby
colors = colorize( tbl.row( field ), path, ax )
for x, y, c, s in zip( xcoords, ycoords, colors, shapes ):
ax.scatter( x, y, color=c, marker=s )
ax.set_xlim( min( xcoords ), max( xcoords ) )
ax.set_ylim( min( ycoords ), max( ycoords ) )
mu.funcMargin( ax )
title = path2name( args.table ) if args.title is None else args.title
ax.set_title( "%s | Goodness of fit = %.3f" % ( title, goodness ) )
ax.set_xlabel( "Dimension 1 (%.1f%%)" % (100 * varexp[0] ) )
ax.set_ylabel( "Dimension 2 (%.1f%%)" % (100 * varexp[1] ) )
mu.funcSetTickParams( ax )
for m in args.level_biplot:
level_biplot( ax, embedding, tbl.row( m ) )
for m in args.quant_biplot:
quant_biplot( ax, embedding, tbl.row( m ), m )
# Shrink current axis by 20%
box = ax.get_position( )
ax.set_position( [box.x0, box.y0, box.width * 0.7, box.height] )
ax.legend( scatterpoints=1,
fontsize=8,
loc='center left',
bbox_to_anchor=(1, 0.5),
)
#plt.tight_layout( )
plt.savefig( args.outfile )
def main():
args = get_args()
T = table(args.table)
if args.select is not None:
T.select(args.select[0], args.select[1], transposed=True)
groups = T.row(args.groups)
if args.last_metadata is not None:
m, T = T.metasplit(args.last_metadata)
T.float()
adonis(T, groups)
def main( ):
args = get_args( )
t = table( args.table, verbose=False )
if args.stratify_by is None:
dt = {}
dt["ALL_SAMPLES"] = t
else:
dt = t.stratify( args.stratify_by )
reports = {}
for level, t in dt.items( ):
if args.last_metadata is not None:
m, t = t.metasplit( args.last_metadata )
reports[level] = check( t )
for level, [summary, failed] in reports.items( ):
print "Considering table <{}> at level <{}>:".format( args.table, level )
print " Summary:", summary
print " FAILED: ", len( failed )
for s, msg in failed.items( ):
print "\t".join( [" FAILED:", s, msg] )
return None
#! usr/bin/env python
import sys
from zopy.table2 import table
t = table(sys.argv[1])
t.float()
t.groupby(lambda x: x[0].split("|")[0], sum)
t.dump()
)
parser.add_argument(
"-l", "--legacy",
action="store_true",
help="iteratively merge tables (better maintains feature order)",
)
args = parser.parse_args()
# ---------------------------------------------------------------
# load and process data
# ---------------------------------------------------------------
p = args.tables[0]
if len( args.tables ) == 1:
t = table( p )
elif args.legacy:
t = table( p )
for p2 in args.tables[1:]:
t2 = table( p2 )
t.merge( t2 )
else:
data = {}
for p in args.tables:
d = table( p ).table2nesteddict( )
for r in d:
inner = data.setdefault( r, {} )
for c in d[r]:
if c in inner and inner[c] != d[r][c]:
warn( p, "overwrites", r, c, inner[c], "with", d[r][c] )
inner[c] = d[r][c]
f: focus on non-header field (first arg) i: invert m: protect/reattach metadata x: read choices from file (one per line) =============================================== Author: Eric Franzosa ([email protected]) """ import os, sys from zopy.table2 import table args = sys.argv[1:] # first arg is a command cluster; read stdin if args[0][0] == "-": t = table() # interpret first arg as a file name else: t = table(args[0]) args = args[1:] # determine operations ops = [] i = 0 while i < len(args): if args[i][0] == "-": op = [args[i], []] i += 1 while i < len(args) and args[i][0] != "-": op[1].append(args[i]) i += 1
strInputPath = args.input
strMode = args.mode
strOutputPath = args.output if args.output is not None else ".".join( strInputPath, strMode )
fReadLength = args.read_length
strSampleReadsPath = args.sample_reads
# ---------------------------------------------------------------
# manipulate data
# ---------------------------------------------------------------
if strMode == "rpkm":
if strSampleReadsPath is None:
sys.exit( "to compute rpkm you must include a file mapping sample IDs to #s of reads" )
dictMillions = col2dict( strSampleReadsPath, key=0, value=1, func=lambda x: float( x ) / 1e6 )
tableCladeRPK = table( strInputPath )
tableCladeRPK.grep( "headers", "s__" )
tableCladeRPK.float()
tableCladeRPK.groupby( lambda x: x.split( "|" )[0], median )
if strMode != "rpk":
for bug, sample, value in tableCladeRPK.iter_entries():
if strMode == "coverage":
tableCladeRPK.set( bug, sample, value * fReadLength / 1e3 )
elif strMode == "rpkm":
tableCladeRPK.set( bug, sample, value / dictMillions[sample] if dictMillions[sample] > 0 else 0 )
tableCladeRPK.unfloat()
tableCladeRPK.colsort()
tableCladeRPK.rowsort()
tableCladeRPK.dump( strOutputPath )
def trunc_normal(m, sd, zmax):
outlier = True
while outlier:
sim = normal(m, sd)
if abs(m - sim) / sd < zmax:
outlier = False
return sim
#-------------------------------------------------------------------------------
# munge hmp data
#-------------------------------------------------------------------------------
zu.say(args.basename, "->", "parsing HMP data")
T = table(args.hmp)
T.select("STSite", args.site, transposed=True)
T.select("VISNO", "1", transposed=True)
T.head("SRS", invert=True)
T.apply_rowheads(lambda x: x.split("|")[-1])
T.grep("headers", "s__")
T.grep("headers", "_unclassified", invert=True)
T.dump("subset.tmp")
T.float()
T.unrarify(1e-20, 1)
bugs = []
for bug, row in T.iter_rows():
stats = []
nonzero = [log(k) / log(10) for k in row if k > 0]
stats.append(len(nonzero) / float(len(row)))
default=0,
type=int,
help="position post splitting to isolate")
args = parser.parse_args()
# process map
idmap, isnewid = {}, {}
with open(args.map) as fh:
for line in fh:
oldid, newid = line.strip().split("\t")
if newid in isnewid:
sys.exit("ERROR: %s listed 2+ times in map" % (newid))
else:
idmap[oldid] = newid
isnewid[newid] = 1
# execute
def applier(colhead):
oldid = colhead.split(args.split)[args.pos]
if oldid in idmap:
return idmap[oldid]
else:
print >> sys.stderr, "can't convert", colhead, "->", oldid
return oldid
t = table(args.table)
t.apply_colheads(applier)
t.dump()
from zopy.table2 import table
from time import time
start = time()
def interval():
global start
stop = time()
print stop - start, "seconds"
start = stop
print "loading"
t1 = table(sys.argv[1])
t2 = table(sys.argv[2])
interval()
print "testing merge"
t1.merge(t2)
interval()
print "loading"
t3 = table(sys.argv[1])
t4 = table(sys.argv[2])
interval()
print "testing merge_old"
t3.merge_old(t4)
interval()
for i in range(len(t1.data)):
default=None,
help="value to insert in place of missing values",
)
parser.add_argument(
"-m",
"--metamerge",
action="store_true",
help="specify if second table is a metadata table",
)
args = parser.parse_args()
# ---------------------------------------------------------------
# load and process data
# ---------------------------------------------------------------
table1 = table(args.input1)
table2 = table(args.input2)
if args.metamerge:
table1.metamerge(table2)
else:
table1.merge(table2)
if args.fill_empty is not None:
table1.apply_entries(lambda x: x if x != c_strNA else args.fill_empty)
# ---------------------------------------------------------------
# dump table
# ---------------------------------------------------------------
# not ideal
if args.output is not None:
table1.dump(args.output)
def main():
args = get_args()
t = table(args.table)
t.float()
# apply scaling
gmax = max([max(row) for header, row in t.iter_rows()])
t.apply_entries(lambda x: (x / gmax)**(1 / float(args.scale)))
# format labels
xlab = sorted(t.colheads)
ylab = sorted(t.rowheads, key=lambda x: argmax(t.rowdict(x)), reverse=True)
# make plot
colors = mu.ncolors(len(xlab), args.colormap)
fig = plt.figure()
axes = mu.funcPlotMatrix([1], [len(t.rowheads), 2])
ax = axes[0][0]
legend = axes[1][0]
ax.xaxis.tick_top()
ax.yaxis.tick_right()
ax.set_xticks(range(len(xlab)))
ax.set_yticks(range(len(ylab)))
ax.set_xticklabels(xlab, rotation=45, rotation_mode="anchor", ha="left")
ax.set_yticklabels(ylab)
ax.set_xlim(-0.5, len(xlab) - 0.5)
ax.set_ylim(-0.5, len(ylab) - 0.5)
# add tiles
for i, x in enumerate(xlab):
for j, y in enumerate(ylab):
value = t.entry(y, x)
if value == 0:
continue
tileh = 1.0 if args.mode == "width" else value
tilew = 1.0 if args.mode == "height" else value
alpha = 1.0
if args.emphasis == "fade" and argmax(t.rowdict(y)) != x:
alpha = c_fadeout
ax.add_patch(
patches.Rectangle(
(i - tilew / 2.0, j - tileh / 2.0),
tilew,
tileh,
edgecolor="white",
facecolor=colors[i],
alpha=alpha,
))
if args.emphasis == "dot" and argmax(t.rowdict(y)) == x:
ax.scatter([i], [j],
s=10,
color="white",
edgecolor="none",
zorder=2)
# add grid
if args.grid == "box":
mu.funcGrid2(
ax,
h=[k - 0.5 for k in range(1, len(ylab))],
v=[k - 0.5 for k in range(1, len(xlab))],
color="0.9",
zorder=2,
border=True,
)
elif args.grid == "cross":
mu.funcGrid2(
ax,
h=[k for k in range(len(ylab))],
v=[k for k in range(len(xlab))],
color="0.9",
zorder=0,
border=True,
)
mu.funcHideBorder(ax)
# cleanup
mu.funcHideTicks(ax)
# draw the legend
mu.funcSetTickParams(legend)
#mu.funcHideBorder( legend )
legend.yaxis.tick_right()
legend.set_yticks([0])
legend.set_yticklabels(["Scale"])
legend.set_xlim(-0.5, len(xlab) - 0.5)
legend.set_xticks(range(len(xlab)))
legend.set_ylim(-0.5, 0.5)
start = nearest_power(gmax, 10)
values = [(10**start) * 0.1**i for i in range(len(xlab))]
values = [(gmax) * 0.1**i for i in range(len(xlab))]
values.reverse()
legend.set_xticklabels(["%.2g" % k for k in values])
j = 0
for i, value in enumerate(values):
value = (value / gmax)**(1 / float(args.scale))
tileh = 1.0 if args.mode == "width" else value
tilew = 1.0 if args.mode == "height" else value
legend.add_patch(
patches.Rectangle(
(i - tilew / 2.0, j - tileh / 2.0),
tilew,
tileh,
edgecolor="none",
facecolor="0.5",
))
# finalize
fig.set_size_inches(
c_colw * len(xlab) + c_rowlabw,
c_rowh * (1 + len(ylab)) + c_collabh,
)
plt.tight_layout()
plt.savefig(args.outfile)
#! /usr/bin/env python import os, sys, re, glob, argparse from zopy.table2 import table from zopy.dictation import col2dict from zopy.utils import path2name dictMap = col2dict( sys.argv[1], key=0, value=1 ) tableData = table( sys.argv[2] ) tableData.apply_colheads( lambda x: path2name( x ) ) tableData.apply_colheads( lambda x: x.split( "." )[0] ) tableData.apply_colheads( lambda x: dictMap[x] ) tableData.dump( sys.argv[3] )
c_strColor2 = "0.60"
c_strBackgroundColor = "0.95"
# constants from args
c_pathPCL = args.input
c_astrSamples = funcArgPathCheck(args.samples)
c_astrClades = funcArgPathCheck(args.clades)
c_strScaling = args.scaling
c_afBins = funcArgPathCheck(args.bins)
if c_afBins is not None:
c_afBins = [float(k) for k in c_afBins]
c_afBins.sort()
c_astrHighlight = funcArgPathCheck(args.highlight)
# load and manipulate table
tableData = table(c_pathPCL)
tableData.float()
if c_astrClades is not None:
tableData.grep("headers", c_astrClades)
if c_astrSamples is not None:
tableData.select("headers", c_astrSamples, transposed=True)
dictCladeTables = tableData.groupify(
lambda strRowhead: strRowhead.split("|")[0])
# override these lists based on what was in the table
c_astrClades = dictCladeTables.keys()
c_astrSamples = tableData.colheads[:]
# derived constants
c_iClades = len(c_astrClades)
c_iSamples = len(c_astrSamples)
# argparse
# ---------------------------------------------------------------
parser = argparse.ArgumentParser()
parser.add_argument('--input', help='')
parser.add_argument('--output', help='')
parser.add_argument('--logmin', type=float, help='')
parser.add_argument('--logmax', type=float, help='')
args = parser.parse_args()
# ---------------------------------------------------------------
# main
# ---------------------------------------------------------------
# get and manipulate data
t = table(args.input)
t.apply_entries(lambda x: None if x == "" else x)
t.apply_entries(lambda x: float(x) if x is not None else None)
# derived features
aColors = [
plt.cm.Dark2(i / float(len(t.colheads)))
for i, colhead in enumerate(t.colheads)
]
aaData = [row for rowhead, row in t.iter_rows()]
logmin, logmax = args.logmin, args.logmax
# make plot
fig = plt.figure()
axes = plt.subplot(111)
stepplot(axes, aaData, colors=aColors)
