def main(args): log = logging.getLogger('root') hdlr = logging.StreamHandler(sys.stdout) log.addHandler(hdlr) log.setLevel(logging.getLevelName(args.loglevel.upper())) dfs = [metadata.parse_fx_par(fn) for fn in args.input] n = dfs[0].shape[0] if not np.all(np.array([df.shape[0] for df in dfs]) == n): log.error("Input files are not aligned!") return 1 df = pd.concat(dfs, axis=0, ignore_index=True) df["CLASS"] = np.repeat(np.arange(1, len(dfs) + 1), n) if args.min_occ: df = df[df["OCC"] >= args.min_occ] df = df.sort_values(by="OCC") df = df.drop_duplicates("C", keep="last") df = df.sort_values(by="C") df = metadata.par2star(df, data_path=args.stack, apix=args.apix, cs=args.cs, ac=args.ac, kv=args.voltage, invert_eulers=args.relion) if args.cls is not None: df = star.select_classes(df, args.cls) star.write_star(args.output, df, reindex=True) return 0
def main(args): data = [] if os.path.isdir(args.input[0]): flist = glob.glob(os.path.join(args.input[0], "*_ctfEstimation.txt")) else: flist = args.input for fn in flist: row = {} with open(fn, 'r') as f: lines = f.readlines() g = lines[1].lstrip("# Input file:").split(" ;")[0] if args.apix is None: args.apix = float(lines[2].lstrip("# Pixel size:").split("Angstrom")[0]) tok = lines[-1].split() if args.path is None: row[star.Relion.MICROGRAPH_NAME] = g else: row[star.Relion.MICROGRAPH_NAME] = os.path.join(args.path, os.path.basename(g)) row[star.Relion.DEFOCUSU] = float(tok[1]) row[star.Relion.DEFOCUSV] = float(tok[2]) row[star.Relion.DEFOCUSANGLE] = float(tok[3]) row[star.Relion.PHASESHIFT] = float(tok[4]) row[star.Relion.CTFFIGUREOFMERIT] = float(tok[5]) row[star.Relion.CTFMAXRESOLUTION] = float(tok[6]) row[star.Relion.MAGNIFICATION] = 10000 row[star.Relion.DETECTORPIXELSIZE] = args.apix data.append(row) df = pd.DataFrame(data) if not args.no_sort: df = star.sort_records(df, inplace=True) df = star.sort_fields(df, inplace=True) star.write_star(args.output, df) return 0
def main(args): log = logging.getLogger('root') hdlr = logging.StreamHandler(sys.stdout) log.addHandler(hdlr) log.setLevel(logging.getLevelName(args.loglevel.upper())) if args.input[0].endswith(".cs"): log.debug("Detected CryoSPARC 2+ .cs file") cs = np.load(args.input[0]) try: df = metadata.parse_cryosparc_2_cs(cs, passthroughs=args.input[1:], minphic=args.minphic, boxsize=args.boxsize, swapxy=args.swapxy, invertx=args.invertx, inverty=args.inverty) except (KeyError, ValueError) as e: log.error(e, exc_info=True) log.error("Required fields could not be mapped. Are you using the right input file(s)?") return 1 else: log.debug("Detected CryoSPARC 0.6.5 .csv file") if len(args.input) > 1: log.error("Only one file at a time supported for CryoSPARC 0.6.5 .csv format") return 1 meta = metadata.parse_cryosparc_065_csv(args.input[0]) # Read cryosparc metadata file. df = metadata.cryosparc_065_csv2star(meta, args.minphic) if args.cls is not None: df = star.select_classes(df, args.cls) if args.copy_micrograph_coordinates is not None: df = star.augment_star_ucsf(df, inplace=True) coord_star = pd.concat( (star.parse_star(inp, keep_index=False, augment=True) for inp in glob(args.copy_micrograph_coordinates)), join="inner") key = star.merge_key(df, coord_star) log.debug("Coordinates merge key: %s" % key) if args.cached or key == star.Relion.IMAGE_NAME: fields = star.Relion.MICROGRAPH_COORDS else: fields = star.Relion.MICROGRAPH_COORDS + [star.UCSF.IMAGE_INDEX, star.UCSF.IMAGE_PATH] df = star.smart_merge(df, coord_star, fields=fields, key=key) star.simplify_star_ucsf(df) if args.micrograph_path is not None: df = star.replace_micrograph_path(df, args.micrograph_path, inplace=True) if args.transform is not None: r = np.array(json.loads(args.transform)) df = star.transform_star(df, r, inplace=True) df = star.check_defaults(df, inplace=True) if args.relion2: df = star.remove_new_relion31(df, inplace=True) star.write_star(args.output, df, resort_records=True, optics=False) else: df = star.remove_deprecated_relion2(df, inplace=True) star.write_star(args.output, df, resort_records=True, optics=True) log.info("Output fields: %s" % ", ".join(df.columns)) return 0
def main(args): star = parse_star(args.input, keep_index=False) if args.cls is not None: clsfields = [f for f in star.columns if "ClassNumber" in f] if len(clsfields) == 0: print("No class labels found") return 1 ind = star[clsfields[0]].isin(args.cls) if not np.any(ind): print("Specified class has no members") return 1 star = star.loc[ind] if args.max_astigmatism is not None: astigmatism = star["rlnDefocusU"] - star["rlnDefocusV"] ind = astigmatism <= args.max_astigmatism star = star.loc[ind] if args.max_resolution is not None: ind = star["rlnFinalResolution"] <= args.max_resolution star = star.loc[ind] if args.min_ctf_fom is not None: ind = star["rlnCtfFigureOfMerit"] >= args.min_ctf_fom star = star.loc[ind] if args.subsample is not None: if args.subsample < 1: args.subsample = np.max(np.round(args.subsample * star.shape[0]), 1) star = star.sample(np.int(args.subsample), random_state=args.seed) write_star(args.output, star) return 0
def main(args): log = logging.getLogger('root') hdlr = logging.StreamHandler(sys.stdout) log.addHandler(hdlr) log.setLevel(logging.getLevelName(args.loglevel.upper())) if args.input.endswith(".cs"): log.debug("Detected CryoSPARC 2+ .cs file") cs = np.load(args.input) try: df = metadata.parse_cryosparc_2_cs(cs, passthrough=args.passthrough, minphic=args.minphic) except (KeyError, ValueError) as e: log.error(e.message) log.error( "A passthrough file may be required (check inside the cryoSPARC 2+ job directory)" ) log.debug(e, exc_info=True) return 1 else: log.debug("Detected CryoSPARC 0.6.5 .csv file") meta = metadata.parse_cryosparc_065_csv( args.input) # Read cryosparc metadata file. df = metadata.cryosparc_065_csv2star(meta, args.minphic) if args.cls is not None: df = star.select_classes(df, args.cls) if args.copy_micrograph_coordinates is not None: coord_star = pd.concat( (star.parse_star(inp, keep_index=False) for inp in glob(args.copy_micrograph_coordinates)), join="inner") star.augment_star_ucsf(coord_star) star.augment_star_ucsf(df) key = star.merge_key(df, coord_star) log.debug("Coordinates merge key: %s" % key) if args.cached or key == star.Relion.IMAGE_NAME: fields = star.Relion.MICROGRAPH_COORDS else: fields = star.Relion.MICROGRAPH_COORDS + [ star.UCSF.IMAGE_INDEX, star.UCSF.IMAGE_PATH ] df = star.smart_merge(df, coord_star, fields=fields, key=key) star.simplify_star_ucsf(df) if args.micrograph_path is not None: df = star.replace_micrograph_path(df, args.micrograph_path, inplace=True) if args.transform is not None: r = np.array(json.loads(args.transform)) df = star.transform_star(df, r, inplace=True) # Write Relion .star file with correct headers. star.write_star(args.output, df, reindex=True) log.info("Output fields: %s" % ", ".join(df.columns)) return 0
def main(args): df = parse_star(args.input, keep_index=False) if args.cls is not None: clsfields = [f for f in df.columns if "ClassNumber" in f] if len(clsfields) == 0: print("No class labels found") return 1 ind = df[clsfields[0]].isin(args.cls) if not np.any(ind): print("Specified class has no members") return 1 df = df.loc[ind] if args.max_astigmatism is not None: astigmatism = df["rlnDefocusU"] - df["rlnDefocusV"] ind = astigmatism <= args.max_astigmatism df = df.loc[ind] if args.max_resolution is not None: if "rlnFinalResolution" in df.columns: ind = df["rlnFinalResolution"] <= args.max_resolution elif "rlnCtfMaxResolution" in df.columns: ind = df["rlnCtfMaxResolution"] <= args.max_resolution else: print("No CTF resolution field found in input") return 1 df = df.loc[ind] if args.max_ctf_fom is not None: ind = df["rlnCtfFigureOfMerit"] <= args.max_ctf_fom df = df.loc[ind] if args.min_ctf_fom is not None: ind = df["rlnCtfFigureOfMerit"] >= args.min_ctf_fom df = df.loc[ind] if args.min_particles is not None: counts = df["rlnMicrographName"].value_counts() subset = df.set_index("rlnMicrographName").loc[counts.index[counts > args.min_particles]] df = subset.reset_index() if args.subsample is not None: if args.subsample < 1: args.subsample = np.max(np.round(args.subsample * df.shape[0]), 1) if args.bootstrap is not None: print("Not implemented yet") return 1 inds = np.random.choice(df.shape[0], size=(np.int(args.subsample), df.shape[0]/np.int(args.subsample)), replace=True) else: df = df.sample(np.int(args.subsample), random_state=args.seed) write_star(args.output, df) return 0
def main(args): log = logging.getLogger('root') hdlr = logging.StreamHandler(sys.stdout) log.addHandler(hdlr) log.setLevel(logging.getLevelName(args.loglevel.upper())) # apix = args.apix = hdr["xlen"] / hdr["nx"] for fn in args.input: if not (fn.endswith(".star") or fn.endswith(".mrcs") or fn.endswith(".mrc")): log.error("Only .star, .mrc, and .mrcs files supported") return 1 first_ptcl = 0 dfs = [] with mrc.ZSliceWriter(args.output) as writer: for fn in args.input: if fn.endswith(".star"): df = star.parse_star(fn, keep_index=False) star.augment_star_ucsf(df) df = df.sort_values([ star.UCSF.IMAGE_ORIGINAL_PATH, star.UCSF.IMAGE_ORIGINAL_INDEX ]) gb = df.groupby(star.UCSF.IMAGE_ORIGINAL_PATH) for name, g in gb: with mrc.ZSliceReader(name) as reader: for i in g[star.UCSF.IMAGE_ORIGINAL_INDEX].values: writer.write(reader.read(i)) else: with mrc.ZSliceReader(fn) as reader: for img in reader: writer.write(img) df = pd.DataFrame( {star.UCSF.IMAGE_ORIGINAL_INDEX: np.arange(reader.nz)}) df[star.UCSF.IMAGE_ORIGINAL_PATH] = fn if args.star is not None: df[star.UCSF.IMAGE_INDEX] = np.arange(first_ptcl, first_ptcl + df.shape[0]) df[star.UCSF.IMAGE_PATH] = writer.path df["index"] = df[star.UCSF.IMAGE_INDEX] star.simplify_star_ucsf(df) dfs.append(df) first_ptcl += df.shape[0] if args.star is not None: df = pd.concat(dfs, join="inner") # df = pd.concat(dfs) # df = df.dropna(df, axis=1, how="any") star.write_star(args.star, df, reindex=True) return 0
def main(args): if args.data_path is None: args.data_path = args.input.rstrip(".par") + ".mrcs" print("No stack path provided, using %s" % args.data_path) df = parse_fx_par(args.input) df = par2star(df, data_path=args.data_path, apix=args.apix, cs=args.cs, ac=args.ac, kv=args.voltage) write_star(args.output, df, reindex=True) return 0
def WriteStarFile(star_file, df): """ Writes a star file from given pandas dataframe. Args: Star file destination, pandas dataframe """ return star.write_star(star_file, df)
def main(args): par = [parse_par(inp) for inp in args.input] if len(par) > 1: idx = par[:, :, "OCC"].idxmax(axis=1) else: par = par[0] if args.data_path is not None: par["Input particle images"] = args.data_path if args.voltage is not None: par["Beam energy (keV)"] = args.voltage if args.cs is not None: par["Spherical aberration (mm)"] = args.cs if args.ac is not None: par["Amplitude contrast"] = args.ac if args.apix is not None: par["Pixel size of images (A)"] = args.apix star = par2star(par) write_star(args.output, star, reindex=True) return 0
def main(args): pyfftw.interfaces.cache.enable() refmap = mrc.read(args.key, compat="relion") df = star.parse_star(args.input, keep_index=False) star.augment_star_ucsf(df) refmap_ft = vop.vol_ft(refmap, threads=args.threads) apix = star.calculate_apix(df) sz = refmap_ft.shape[0] // 2 - 1 sx, sy = np.meshgrid(rfftfreq(sz), fftfreq(sz)) s = np.sqrt(sx**2 + sy**2) r = s * sz r = np.round(r).astype(np.int64) r[r > sz // 2] = sz // 2 + 1 a = np.arctan2(sy, sx) def1 = df["rlnDefocusU"].values def2 = df["rlnDefocusV"].values angast = df["rlnDefocusAngle"].values phase = df["rlnPhaseShift"].values kv = df["rlnVoltage"].values ac = df["rlnAmplitudeContrast"].values cs = df["rlnSphericalAberration"].values xshift = df["rlnOriginX"].values yshift = df["rlnOriginY"].values score = np.zeros(df.shape[0]) # TODO parallelize for i, row in df.iterrows(): xcor = particle_xcorr(row, refmap_ft) if args.top is None: args.top = df.shape[0] top = df.iloc[np.argsort(score)][:args.top] star.simplify_star_ucsf(top) star.write_star(args.output, top) return 0
def main(args): if args.input.endswith(".cs"): cs = np.load(args.input) if args.passthrough is None: if u"blob/path" not in cs.dtype.names: print( "A passthrough file is required (found inside the cryoSPARC 2+ job directory)" ) return 1 df = metadata.parse_cryosparc_2_cs(cs, passthrough=args.passthrough, minphic=args.minphic) else: meta = metadata.parse_cryosparc_065_csv( args.input) # Read cryosparc metadata file. df = metadata.cryosparc_065_csv2star(meta, args.minphic) if args.cls is not None: df = star.select_classes(df, args.cls) if args.copy_micrograph_coordinates is not None: coord_star = pd.concat( (star.parse_star(inp, keep_index=False) for inp in glob(args.copy_micrograph_coordinates)), join="inner") df = star.smart_merge(df, coord_star, fields=star.Relion.MICROGRAPH_COORDS) if args.transform is not None: r = np.array(json.loads(args.transform)) df = star.transform_star(df, r, inplace=True) # Write Relion .star file with correct headers. star.write_star(args.output, df, reindex=True) return 0
def main(args): if args.info: args.input.append(args.output) df = pd.concat( (star.parse_star(inp, augment=args.augment) for inp in args.input), join="inner") dfaux = None if args.cls is not None: df = star.select_classes(df, args.cls) if args.info: if star.is_particle_star(df) and star.Relion.CLASS in df.columns: c = df[star.Relion.CLASS].value_counts() print("%s particles in %d classes" % ("{:,}".format(df.shape[0]), len(c))) print(" ".join([ '%d: %s (%.2f %%)' % (i, "{:,}".format(s), 100. * s / c.sum()) for i, s in c.sort_index().iteritems() ])) elif star.is_particle_star(df): print("%s particles" % "{:,}".format(df.shape[0])) if star.Relion.MICROGRAPH_NAME in df.columns: mgraphcnt = df[star.Relion.MICROGRAPH_NAME].value_counts() print( "%s micrographs, %s +/- %s particles per micrograph" % ("{:,}".format(len(mgraphcnt)), "{:,.3f}".format( np.mean(mgraphcnt)), "{:,.3f}".format(np.std(mgraphcnt)))) try: print("%f A/px (%sX magnification)" % (star.calculate_apix(df), "{:,.0f}".format( df[star.Relion.MAGNIFICATION][0]))) except KeyError: pass return 0 if args.drop_angles: df.drop(star.Relion.ANGLES, axis=1, inplace=True, errors="ignore") if args.drop_containing is not None: containing_fields = [ f for q in args.drop_containing for f in df.columns if q in f ] if args.invert: containing_fields = df.columns.difference(containing_fields) df.drop(containing_fields, axis=1, inplace=True, errors="ignore") if args.offset_group is not None: df[star.Relion.GROUPNUMBER] += args.offset_group if args.subsample_micrographs is not None: if args.bootstrap is not None: print("Only particle sampling allows bootstrapping") return 1 mgraphs = df[star.Relion.MICROGRAPH_NAME].unique() if args.subsample_micrographs < 1: args.subsample_micrographs = np.int( max(np.round(args.subsample_micrographs * len(mgraphs)), 1)) else: args.subsample_micrographs = np.int(args.subsample_micrographs) ind = np.random.choice(len(mgraphs), size=args.subsample_micrographs, replace=False) mask = df[star.Relion.MICROGRAPH_NAME].isin(mgraphs[ind]) if args.auxout is not None: dfaux = df.loc[~mask] df = df.loc[mask] if args.subsample is not None and args.suffix == "": if args.subsample < 1: args.subsample = np.int( max(np.round(args.subsample * df.shape[0]), 1)) else: args.subsample = np.int(args.subsample) ind = np.random.choice(df.shape[0], size=args.subsample, replace=False) mask = df.index.isin(ind) if args.auxout is not None: dfaux = df.loc[~mask] df = df.loc[mask] if args.copy_angles is not None: angle_star = star.parse_star(args.copy_angles, augment=args.augment) df = star.smart_merge(df, angle_star, fields=star.Relion.ANGLES) if args.transform is not None: if args.transform.count(",") == 2: r = star.euler2rot( *np.deg2rad([np.double(s) for s in args.transform.split(",")])) else: r = np.array(json.loads(args.transform)) df = star.transform_star(df, r, inplace=True) if args.invert_hand: df[star.Relion.ANGLEROT] = -df[star.Relion.ANGLEROT] df[star.Relion.ANGLETILT] = 180 - df[star.Relion.ANGLETILT] if args.copy_paths is not None: path_star = star.parse_star(args.copy_paths) df[star.Relion.IMAGE_NAME] = path_star[star.Relion.IMAGE_NAME] if args.copy_ctf is not None: ctf_star = pd.concat((star.parse_star(inp, augment=args.augment) for inp in glob.glob(args.copy_ctf)), join="inner") df = star.smart_merge(df, ctf_star, star.Relion.CTF_PARAMS) if args.copy_micrograph_coordinates is not None: coord_star = pd.concat( (star.parse_star(inp, augment=args.augment) for inp in glob.glob(args.copy_micrograph_coordinates)), join="inner") df = star.smart_merge(df, coord_star, fields=star.Relion.MICROGRAPH_COORDS) if args.scale is not None: star.scale_coordinates(df, args.scale, inplace=True) star.scale_origins(df, args.scale, inplace=True) star.scale_magnification(df, args.scale, inplace=True) if args.scale_particles is not None: star.scale_origins(df, args.scale, inplace=True) star.scale_magnification(df, args.scale, inplace=True) if args.scale_coordinates is not None: star.scale_coordinates(df, args.scale_coordinates, inplace=True) if args.scale_origins is not None: star.scale_origins(df, args.scale_origins, inplace=True) if args.scale_magnification is not None: star.scale_magnification(df, args.scale_magnfication, inplace=True) if args.recenter: df = star.recenter(df, inplace=True) if args.zero_origins: df = star.zero_origins(df, inplace=True) if args.pick: df.drop(df.columns.difference(star.Relion.PICK_PARAMS), axis=1, inplace=True, errors="ignore") if args.subsample is not None and args.suffix != "": if args.subsample < 1: print("Specific integer sample size") return 1 nsamplings = args.bootstrap if args.bootstrap is not None else df.shape[ 0] / np.int(args.subsample) inds = np.random.choice(df.shape[0], size=(nsamplings, np.int(args.subsample)), replace=args.bootstrap is not None) for i, ind in enumerate(inds): star.write_star( os.path.join( args.output, os.path.basename(args.input[0])[:-5] + args.suffix + "_%d" % (i + 1)), df.iloc[ind]) if args.to_micrographs: gb = df.groupby(star.Relion.MICROGRAPH_NAME) mu = gb.mean() df = mu[[ c for c in star.Relion.CTF_PARAMS + star.Relion.MICROSCOPE_PARAMS + [star.Relion.MICROGRAPH_NAME] if c in mu ]].reset_index() if args.micrograph_range: df.set_index(star.Relion.MICROGRAPH_NAME, inplace=True) m, n = [int(tok) for tok in args.micrograph_range.split(",")] mg = df.index.unique().sort_values() outside = list(range(0, m)) + list(range(n, len(mg))) dfaux = df.loc[mg[outside]].reset_index() df = df.loc[mg[m:n]].reset_index() if args.micrograph_path is not None: df = star.replace_micrograph_path(df, args.micrograph_path, inplace=True) if args.min_separation is not None: gb = df.groupby(star.Relion.MICROGRAPH_NAME) dupes = [] for n, g in gb: nb = algo.query_connected( g[star.Relion.COORDS], args.min_separation / star.calculate_apix(df)) dupes.extend(g.index[~np.isnan(nb)]) dfaux = df.loc[dupes] df.drop(dupes, inplace=True) if args.merge_source is not None: if args.merge_fields is not None: if "," in args.merge_fields: args.merge_fields = args.merge_fields.split(",") else: args.merge_fields = [args.merge_fields] else: print("Merge fields must be specified using --merge-fields") return 1 if args.merge_key is not None: if "," in args.merge_key: args.merge_key = args.merge_key.split(",") merge_star = star.parse_star(args.merge_source, augment=args.augment) df = star.smart_merge(df, merge_star, fields=args.merge_fields, key=args.merge_key) if args.split_micrographs: dfs = star.split_micrographs(df) for mg in dfs: star.write_star( os.path.join(args.output, os.path.basename(mg)[:-4]) + args.suffix, dfs[mg]) return 0 if args.auxout is not None and dfaux is not None: star.write_star(args.auxout, dfaux, simplify=args.augment) if args.output is not None: star.write_star(args.output, df, simplify=args.augment) return 0
def main(args): log = logging.getLogger(__name__) log.setLevel(logging.INFO) hdlr = logging.StreamHandler(sys.stdout) if args.quiet: hdlr.setLevel(logging.WARNING) else: hdlr.setLevel(logging.INFO) log.addHandler(hdlr) if args.target is None and args.sym is None: log.error( "At least a target or symmetry group must be provided via --target or --sym" ) return 1 elif args.target is not None and args.boxsize is None and args.origin is None: log.error("An origin must be provided via --boxsize or --origin") return 1 if args.target is not None: try: args.target = np.array( [np.double(tok) for tok in args.target.split(",")]) except: log.error( "Target must be comma-separated list of x,y,z coordinates") return 1 if args.origin is not None: if args.boxsize is not None: log.warn("--origin supersedes --boxsize") try: args.origin = np.array( [np.double(tok) for tok in args.origin.split(",")]) except: log.error( "Origin must be comma-separated list of x,y,z coordinates") return 1 if args.sym is not None: args.sym = util.relion_symmetry_group(args.sym) df = star.parse_star(args.input) if args.apix is None: args.apix = star.calculate_apix(df) if args.apix is None: log.warn( "Could not compute pixel size, default is 1.0 Angstroms per pixel" ) args.apix = 1.0 df[star.Relion.MAGNIFICATION] = 10000 df[star.DETECTORPIXELSIZE] = 1.0 if args.cls is not None: df = star.select_classes(df, args.cls) if args.target is not None: if args.origin is not None: args.origin /= args.apix elif args.boxsize is not None: args.origin = np.ones(3) * args.boxsize / 2 args.target /= args.apix c = args.target - args.origin c = np.where(np.abs(c) < 1, 0, c) # Ignore very small coordinates. d = np.linalg.norm(c) ax = c / d cm = util.euler2rot(*np.array( [np.arctan2(ax[1], ax[0]), np.arccos(ax[2]), np.deg2rad(args.psi)])) ops = [op.dot(cm) for op in args.sym] if args.sym is not None else [cm] dfs = [ star.transform_star(df, op.T, -d, rotate=args.shift_only, invert=args.target_invert, adjust_defocus=args.adjust_defocus) for op in ops ] elif args.sym is not None: dfs = list( subparticle_expansion(df, args.sym, -args.displacement / args.apix)) else: log.error( "At least a target or symmetry group must be provided via --target or --sym" ) return 1 if args.recenter: for s in dfs: star.recenter(s, inplace=True) if args.suffix is None and not args.skip_join: if len(dfs) > 1: df = util.interleave(dfs) else: df = dfs[0] star.write_star(args.output, df) else: for i, s in enumerate(dfs): star.write_star(os.path.join(args.output, args.suffix + "_%d" % i), s) return 0
def main(args): log = logging.getLogger(__name__) hdlr = logging.StreamHandler(sys.stdout) log.addHandler(hdlr) log.setLevel(logging.getLevelName(args.loglevel.upper())) if args.target is None and args.sym is None and args.transform is None and args.euler is None: log.error("At least a target, transformation matrix, Euler angles, or a symmetry group must be provided") return 1 elif (args.target is not None or args.transform is not None) and args.boxsize is None and args.origin is None: log.error("An origin must be provided via --boxsize or --origin") return 1 if args.apix is None: df = star.parse_star(args.input, nrows=1) args.apix = star.calculate_apix(df) if args.apix is None: log.warn("Could not compute pixel size, default is 1.0 Angstroms per pixel") args.apix = 1.0 df[star.Relion.MAGNIFICATION] = 10000 df[star.Relion.DETECTORPIXELSIZE] = 1.0 if args.target is not None: try: args.target = np.array([np.double(tok) for tok in args.target.split(",")]) except: log.error("Target must be comma-separated list of x,y,z coordinates") return 1 if args.euler is not None: try: args.euler = np.deg2rad(np.array([np.double(tok) for tok in args.euler.split(",")])) args.transform = np.zeros((3, 4)) args.transform[:, :3] = geom.euler2rot(*args.euler) if args.target is not None: args.transform[:, -1] = args.target except: log.error("Euler angles must be comma-separated list of rotation, tilt, skew in degrees") return 1 if args.transform is not None and not hasattr(args.transform, "dtype"): if args.target is not None: log.warn("--target supersedes --transform") try: args.transform = np.array(json.loads(args.transform)) except: log.error("Transformation matrix must be in JSON/Numpy format") return 1 if args.origin is not None: if args.boxsize is not None: log.warn("--origin supersedes --boxsize") try: args.origin = np.array([np.double(tok) for tok in args.origin.split(",")]) args.origin /= args.apix except: log.error("Origin must be comma-separated list of x,y,z coordinates") return 1 elif args.boxsize is not None: args.origin = np.ones(3) * args.boxsize / 2 if args.sym is not None: args.sym = util.relion_symmetry_group(args.sym) df = star.parse_star(args.input) if star.calculate_apix(df) != args.apix: log.warn("Using specified pixel size of %f instead of calculated size %f" % (args.apix, star.calculate_apix(df))) if args.cls is not None: df = star.select_classes(df, args.cls) if args.target is not None: args.target /= args.apix c = args.target - args.origin c = np.where(np.abs(c) < 1, 0, c) # Ignore very small coordinates. d = np.linalg.norm(c) ax = c / d r = geom.euler2rot(*np.array([np.arctan2(ax[1], ax[0]), np.arccos(ax[2]), np.deg2rad(args.psi)])) d = -d elif args.transform is not None: r = args.transform[:, :3] if args.transform.shape[1] == 4: d = args.transform[:, -1] / args.apix d = r.dot(args.origin) + d - args.origin else: d = 0 elif args.sym is not None: r = np.identity(3) d = -args.displacement / args.apix else: log.error("At least a target or symmetry group must be provided via --target or --sym") return 1 log.debug("Final rotation: %s" % str(r).replace("\n", "\n" + " " * 16)) ops = [op.dot(r.T) for op in args.sym] if args.sym is not None else [r.T] log.debug("Final translation: %s (%f px)" % (str(d), np.linalg.norm(d))) dfs = list(subparticle_expansion(df, ops, d, rotate=args.shift_only, invert=args.invert, adjust_defocus=args.adjust_defocus)) if args.recenter: for s in dfs: star.recenter(s, inplace=True) if args.suffix is None and not args.skip_join: if len(dfs) > 1: df = util.interleave(dfs) else: df = dfs[0] df = star.compatible(df, relion2=args.relion2, inplace=True) star.write_star(args.output, df, optics=(not args.relion2)) else: for i, s in enumerate(dfs): s = star.compatible(s, relion2=args.relion2, inplace=True) star.write_star(os.path.join(args.output, args.suffix + "_%d" % i), s, optics=(not args.relion2)) return 0
def main(args): log = logging.getLogger('root') hdlr = logging.StreamHandler(sys.stdout) log.addHandler(hdlr) log.setLevel(logging.getLevelName(args.loglevel.upper())) df = star.parse_star(args.input, keep_index=False) star.augment_star_ucsf(df) maxshift = np.round(np.max(np.abs(df[star.Relion.ORIGINS].values))) if args.map is not None: if args.map.endswith(".npy"): log.info("Reading precomputed 3D FFT of volume") f3d = np.load(args.map) log.info("Finished reading 3D FFT of volume") if args.size is None: args.size = (f3d.shape[0] - 3) // args.pfac else: vol = mrc.read(args.map, inc_header=False, compat="relion") if args.mask is not None: mask = mrc.read(args.mask, inc_header=False, compat="relion") vol *= mask if args.size is None: args.size = vol.shape[0] if args.crop is not None and args.size // 2 < maxshift + args.crop // 2: log.error( "Some shifts are too large to crop (maximum crop is %d)" % (args.size - 2 * maxshift)) return 1 log.info("Preparing 3D FFT of volume") f3d = vop.vol_ft(vol, pfac=args.pfac, threads=args.threads) log.info("Finished 3D FFT of volume") else: log.error("Please supply a map") return 1 sz = (f3d.shape[0] - 3) // args.pfac apix = star.calculate_apix(df) * np.double(args.size) / sz sx, sy = np.meshgrid(np.fft.rfftfreq(sz), np.fft.fftfreq(sz)) s = np.sqrt(sx**2 + sy**2) a = np.arctan2(sy, sx) log.info("Projection size is %d, unpadded volume size is %d" % (args.size, sz)) log.info("Effective pixel size is %f A/px" % apix) if args.subtract and args.size != sz: log.error("Volume and projections must be same size when subtracting") return 1 if args.crop is not None and args.size // 2 < maxshift + args.crop // 2: log.error("Some shifts are too large to crop (maximum crop is %d)" % (args.size - 2 * maxshift)) return 1 ift = None with mrc.ZSliceWriter(args.output, psz=apix) as zsw: for i, p in df.iterrows(): f2d = project(f3d, p, s, sx, sy, a, pfac=args.pfac, apply_ctf=args.ctf, size=args.size, flip_phase=args.flip) if ift is None: ift = irfft2(f2d.copy(), threads=args.threads, planner_effort="FFTW_ESTIMATE", auto_align_input=True, auto_contiguous=True) proj = fftshift( ift(f2d.copy(), np.zeros(ift.output_shape, dtype=ift.output_dtype))) log.debug("%f +/- %f" % (np.mean(proj), np.std(proj))) if args.subtract: with mrc.ZSliceReader(p["ucsfImagePath"]) as zsr: img = zsr.read(p["ucsfImageIndex"]) log.debug("%f +/- %f" % (np.mean(img), np.std(img))) proj = img - proj if args.crop is not None: orihalf = args.size // 2 newhalf = args.crop // 2 x = orihalf - np.int(np.round(p[star.Relion.ORIGINX])) y = orihalf - np.int(np.round(p[star.Relion.ORIGINY])) proj = proj[y - newhalf:y + newhalf, x - newhalf:x + newhalf] zsw.write(proj) log.debug( "%d@%s: %d/%d" % (p["ucsfImageIndex"], p["ucsfImagePath"], i + 1, df.shape[0])) if args.star is not None: log.info("Writing output .star file") if args.crop is not None: df = star.recenter(df, inplace=True) if args.subtract: df[star.UCSF.IMAGE_ORIGINAL_PATH] = df[star.UCSF.IMAGE_PATH] df[star.UCSF.IMAGE_ORIGINAL_INDEX] = df[star.UCSF.IMAGE_INDEX] df[star.UCSF.IMAGE_PATH] = args.output df[star.UCSF.IMAGE_INDEX] = np.arange(df.shape[0]) star.simplify_star_ucsf(df) star.write_star(args.star, df) return 0
def main(args): if args.info: args.input.append(args.output) df = pd.concat( (star.parse_star(inp, augment=args.augment) for inp in args.input), join="inner") dfaux = None if args.cls is not None: df = star.select_classes(df, args.cls) if args.info: if star.is_particle_star(df) and star.Relion.CLASS in df.columns: c = df[star.Relion.CLASS].value_counts() print("%s particles in %d classes" % ("{:,}".format(df.shape[0]), len(c))) print(" ".join([ '%d: %s (%.2f %%)' % (i, "{:,}".format(s), 100. * s / c.sum()) for i, s in iteritems(c.sort_index()) ])) elif star.is_particle_star(df): print("%s particles" % "{:,}".format(df.shape[0])) if star.Relion.MICROGRAPH_NAME in df.columns: mgraphcnt = df[star.Relion.MICROGRAPH_NAME].value_counts() print( "%s micrographs, %s +/- %s particles per micrograph" % ("{:,}".format(len(mgraphcnt)), "{:,.3f}".format( np.mean(mgraphcnt)), "{:,.3f}".format(np.std(mgraphcnt)))) try: print("%f A/px (%sX magnification)" % (star.calculate_apix(df), "{:,.0f}".format( df[star.Relion.MAGNIFICATION][0]))) except KeyError: pass if len(df.columns.intersection(star.Relion.ORIGINS3D)) > 0: print("Largest shift is %f pixels" % np.max( np.abs(df[df.columns.intersection( star.Relion.ORIGINS3D)].values))) return 0 if args.drop_angles: df.drop(star.Relion.ANGLES, axis=1, inplace=True, errors="ignore") if args.drop_containing is not None: containing_fields = [ f for q in args.drop_containing for f in df.columns if q in f ] if args.invert: containing_fields = df.columns.difference(containing_fields) df.drop(containing_fields, axis=1, inplace=True, errors="ignore") if args.offset_group is not None: df[star.Relion.GROUPNUMBER] += args.offset_group if args.restack is not None: if not args.augment: star.augment_star_ucsf(df, inplace=True) star.set_original_fields(df, inplace=True) df[star.UCSF.IMAGE_PATH] = args.restack df[star.UCSF.IMAGE_INDEX] = np.arange(df.shape[0]) if args.subsample_micrographs is not None: if args.bootstrap is not None: print("Only particle sampling allows bootstrapping") return 1 mgraphs = df[star.Relion.MICROGRAPH_NAME].unique() if args.subsample_micrographs < 1: args.subsample_micrographs = np.int( max(np.round(args.subsample_micrographs * len(mgraphs)), 1)) else: args.subsample_micrographs = np.int(args.subsample_micrographs) ind = np.random.choice(len(mgraphs), size=args.subsample_micrographs, replace=False) mask = df[star.Relion.MICROGRAPH_NAME].isin(mgraphs[ind]) if args.auxout is not None: dfaux = df.loc[~mask] df = df.loc[mask] if args.subsample is not None and args.suffix == "": if args.subsample < 1: args.subsample = np.int( max(np.round(args.subsample * df.shape[0]), 1)) else: args.subsample = np.int(args.subsample) ind = np.random.choice(df.shape[0], size=args.subsample, replace=False) mask = df.index.isin(ind) if args.auxout is not None: dfaux = df.loc[~mask] df = df.loc[mask] if args.copy_angles is not None: angle_star = star.parse_star(args.copy_angles, augment=args.augment) df = star.smart_merge(df, angle_star, fields=star.Relion.ANGLES, key=args.merge_key) if args.copy_alignments is not None: align_star = star.parse_star(args.copy_alignments, augment=args.augment) df = star.smart_merge(df, align_star, fields=star.Relion.ALIGNMENTS, key=args.merge_key) if args.copy_reconstruct_images is not None: recon_star = star.parse_star(args.copy_reconstruct_images, augment=args.augment) df[star.Relion.RECONSTRUCT_IMAGE_NAME] = recon_star[ star.Relion.IMAGE_NAME] if args.transform is not None: if args.transform.count(",") == 2: r = geom.euler2rot( *np.deg2rad([np.double(s) for s in args.transform.split(",")])) else: r = np.array(json.loads(args.transform)) df = star.transform_star(df, r, inplace=True) if args.invert_hand: df = star.invert_hand(df, inplace=True) if args.copy_paths is not None: path_star = star.parse_star(args.copy_paths) star.set_original_fields(df, inplace=True) df[star.Relion.IMAGE_NAME] = path_star[star.Relion.IMAGE_NAME] if args.copy_ctf is not None: ctf_star = pd.concat((star.parse_star(inp, augment=args.augment) for inp in glob.glob(args.copy_ctf)), join="inner") df = star.smart_merge(df, ctf_star, star.Relion.CTF_PARAMS, key=args.merge_key) if args.copy_micrograph_coordinates is not None: coord_star = pd.concat( (star.parse_star(inp, augment=args.augment) for inp in glob.glob(args.copy_micrograph_coordinates)), join="inner") df = star.smart_merge(df, coord_star, fields=star.Relion.MICROGRAPH_COORDS, key=args.merge_key) if args.scale is not None: star.scale_coordinates(df, args.scale, inplace=True) star.scale_origins(df, args.scale, inplace=True) star.scale_magnification(df, args.scale, inplace=True) if args.scale_particles is not None: star.scale_origins(df, args.scale_particles, inplace=True) star.scale_magnification(df, args.scale_particles, inplace=True) if args.scale_coordinates is not None: star.scale_coordinates(df, args.scale_coordinates, inplace=True) if args.scale_origins is not None: star.scale_origins(df, args.scale_origins, inplace=True) if args.scale_magnification is not None: star.scale_magnification(df, args.scale_magnification, inplace=True) if args.scale_apix is not None: star.scale_apix(df, args.scale_apix, inplace=True) if args.recenter: df = star.recenter(df, inplace=True) if args.zero_origins: df = star.zero_origins(df, inplace=True) if args.pick: df.drop(df.columns.difference(star.Relion.PICK_PARAMS), axis=1, inplace=True, errors="ignore") if args.subsample is not None and args.suffix != "": if args.subsample < 1: print("Specific integer sample size") return 1 nsamplings = args.bootstrap if args.bootstrap is not None else df.shape[ 0] / np.int(args.subsample) inds = np.random.choice(df.shape[0], size=(nsamplings, np.int(args.subsample)), replace=args.bootstrap is not None) for i, ind in enumerate(inds): star.write_star( os.path.join( args.output, os.path.basename(args.input[0])[:-5] + args.suffix + "_%d" % (i + 1)), df.iloc[ind]) if args.to_micrographs: df = star.to_micrographs(df) if args.micrograph_range: df.set_index(star.Relion.MICROGRAPH_NAME, inplace=True) m, n = [int(tok) for tok in args.micrograph_range.split(",")] mg = df.index.unique().sort_values() outside = list(range(0, m)) + list(range(n, len(mg))) dfaux = df.loc[mg[outside]].reset_index() df = df.loc[mg[m:n]].reset_index() if args.micrograph_path is not None: df = star.replace_micrograph_path(df, args.micrograph_path, inplace=True) if args.min_separation is not None: gb = df.groupby(star.Relion.MICROGRAPH_NAME) dupes = [] for n, g in gb: nb = algo.query_connected( g[star.Relion.COORDS].values - g[star.Relion.ORIGINS], args.min_separation / star.calculate_apix(df)) dupes.extend(g.index[~np.isnan(nb)]) dfaux = df.loc[dupes] df.drop(dupes, inplace=True) if args.merge_source is not None: if args.merge_fields is not None: if "," in args.merge_fields: args.merge_fields = args.merge_fields.split(",") else: args.merge_fields = [args.merge_fields] else: print("Merge fields must be specified using --merge-fields") return 1 if args.merge_key is not None: if "," in args.merge_key: args.merge_key = args.merge_key.split(",") if args.by_original: args.by_original = star.original_field(args.merge_key) else: args.by_original = args.merge_key merge_star = star.parse_star(args.merge_source, augment=args.augment) df = star.smart_merge(df, merge_star, fields=args.merge_fields, key=args.merge_key, left_key=args.by_original) if args.revert_original: df = star.revert_original(df, inplace=True) if args.set_optics is not None: tok = args.set_optics.split(",") df = star.set_optics_groups(df, sep=tok[0], idx=int(tok[1]), inplace=True) df.dropna(axis=0, how="any", inplace=True) if args.drop_optics_group is not None: idx = df[star.Relion.OPTICSGROUP].isin(args.drop_optics_group) if not np.any(idx): idx = df[star.Relion.OPTICSGROUPNAME].isin(args.drop_optics_group) if not np.any(idx): print("No group found to drop") return 1 df = df.loc[~idx] if args.split_micrographs: dfs = star.split_micrographs(df) for mg in dfs: star.write_star( os.path.join(args.output, os.path.basename(mg)[:-4]) + args.suffix, dfs[mg]) return 0 if args.auxout is not None and dfaux is not None: if not args.relion2: df = star.remove_deprecated_relion2(dfaux, inplace=True) star.write_star(args.output, df, resort_records=args.sort, simplify=args.augment_output, optics=True) else: df = star.remove_new_relion31(dfaux, inplace=True) star.write_star(args.output, df, resort_records=args.sort, simplify=args.augment_output, optics=False) if args.output is not None: if not args.relion2: # Relion 3.1 style output. df = star.remove_deprecated_relion2(df, inplace=True) star.write_star(args.output, df, resort_records=args.sort, simplify=args.augment_output, optics=True) else: df = star.remove_new_relion31(df, inplace=True) star.write_star(args.output, df, resort_records=args.sort, simplify=args.augment_output, optics=False) return 0
def main(args): log = logging.getLogger('root') hdlr = logging.StreamHandler(sys.stdout) log.addHandler(hdlr) log.setLevel(logging.getLevelName(args.loglevel.upper())) df = star.parse_star(args.input, keep_index=False) star.augment_star_ucsf(df) if args.map is not None: vol = mrc.read(args.map, inc_header=False, compat="relion") if args.mask is not None: mask = mrc.read(args.mask, inc_header=False, compat="relion") vol *= mask else: print("Please supply a map") return 1 f3d = vop.vol_ft(vol, pfac=args.pfac, threads=args.threads) sz = f3d.shape[0] // 2 - 1 sx, sy = np.meshgrid(np.fft.rfftfreq(sz), np.fft.fftfreq(sz)) s = np.sqrt(sx**2 + sy**2) a = np.arctan2(sy, sx) ift = None with mrc.ZSliceWriter(args.output) as zsw: for i, p in df.iterrows(): f2d = project(f3d, p, s, sx, sy, a, apply_ctf=args.ctf, size=args.size) if ift is None: ift = irfft2(f2d.copy(), threads=cpu_count(), planner_effort="FFTW_ESTIMATE", auto_align_input=True, auto_contiguous=True) proj = fftshift( ift(f2d.copy(), np.zeros(vol.shape[:-1], dtype=vol.dtype))) log.debug("%f +/- %f" % (np.mean(proj), np.std(proj))) if args.subtract: with mrc.ZSliceReader(p["ucsfImagePath"]) as zsr: img = zsr.read(p["ucsfImageIndex"]) log.debug("%f +/- %f" % (np.mean(img), np.std(img))) proj = img - proj zsw.write(proj) log.info( "%d@%s: %d/%d" % (p["ucsfImageIndex"], p["ucsfImagePath"], i + 1, df.shape[0])) if args.star is not None: if args.subtract: df[star.UCSF.IMAGE_ORIGINAL_PATH] = df[star.UCSF.IMAGE_PATH] df[star.UCSF.IMAGE_ORIGINAL_INDEX] = df[star.UCSF.IMAGE_INDEX] df[star.UCSF.IMAGE_PATH] = args.output df[star.UCSF.IMAGE_INDEX] = np.arange(df.shape[0]) star.simplify_star_ucsf(df) star.write_star(args.star, df) return 0
def main(args): log = logging.getLogger('root') hdlr = logging.StreamHandler(sys.stdout) log.addHandler(hdlr) log.setLevel(logging.getLevelName(args.loglevel.upper())) dfs = [metadata.parse_fx_par(fn) for fn in args.input] n = dfs[0].shape[0] if not np.all(np.array([df.shape[0] for df in dfs]) == n): log.error("Input files are not aligned!") return 1 df = pd.concat(dfs, axis=0, ignore_index=True) df["CLASS"] = np.repeat(np.arange(1, len(dfs) + 1), n) if args.min_occ: df = df[df["OCC"] >= args.min_occ] df = df.sort_values(by="OCC") df = df.drop_duplicates("C", keep="last") df = df.sort_values(by="C") df.reset_index(inplace=True) if args.min_score is not None: if args.min_score < 1: args.min_score = np.percentile(df["SCORE"], (1 - args.min_score) * 100) df = df.loc[df["SCORE"] >= args.min_score] if args.merge is not None: dfo = star.parse_star(args.merge) args.apix = star.calculate_apix(dfo) args.cs = dfo.iloc[0][star.Relion.CS] args.ac = dfo.iloc[0][star.Relion.AC] args.voltage = dfo.iloc[0][star.Relion.VOLTAGE] df = metadata.par2star(df, data_path=args.stack, apix=args.apix, cs=args.cs, ac=args.ac, kv=args.voltage, invert_eulers=args.invert_eulers) if args.stack is None: df[star.UCSF.IMAGE_INDEX] = dfo[star.UCSF.IMAGE_INDEX] df[star.UCSF.IMAGE_PATH] = dfo[star.UCSF.IMAGE_PATH] key = [star.UCSF.IMAGE_INDEX, star.UCSF.IMAGE_PATH] fields = star.Relion.MICROGRAPH_COORDS + [ star.UCSF.IMAGE_ORIGINAL_INDEX, star.UCSF.IMAGE_ORIGINAL_PATH ] + [star.Relion.OPTICSGROUP ] + star.Relion.OPTICSGROUPTABLE + [star.Relion.RANDOMSUBSET] df = star.smart_merge(df, dfo, fields=fields, key=key) if args.revert_original: df = star.revert_original(df, inplace=True) else: df = metadata.par2star(df, data_path=args.stack, apix=args.apix, cs=args.cs, ac=args.ac, kv=args.voltage, invert_eulers=args.invert_eulers) if args.cls is not None: df = star.select_classes(df, args.cls) df = star.check_defaults(df, inplace=True) df = star.compatible(df, relion2=args.relion2, inplace=True) star.write_star(args.output, df, optics=(not args.relion2)) return 0
def main(args): log = logging.getLogger('root') hdlr = logging.StreamHandler(sys.stdout) log.addHandler(hdlr) log.setLevel(logging.getLevelName(args.loglevel.upper())) # apix = args.apix = hdr["xlen"] / hdr["nx"] for fn in args.input: if not (fn.endswith(".star") or fn.endswith(".mrcs") or fn.endswith(".mrc") or fn.endswith(".par")): log.error("Only .star, .mrc, .mrcs, and .par files supported") return 1 first_ptcl = 0 dfs = [] with mrc.ZSliceWriter(args.output) as writer: for fn in args.input: if fn.endswith(".star"): df = star.parse_star(fn, augment=True) if args.cls is not None: df = star.select_classes(df, args.cls) star.set_original_fields(df, inplace=True) if args.resort: df = df.sort_values([star.UCSF.IMAGE_ORIGINAL_PATH, star.UCSF.IMAGE_ORIGINAL_INDEX]) for idx, row in df.iterrows(): if args.stack_path is not None: input_stack_path = os.path.join(args.stack_path, row[star.UCSF.IMAGE_ORIGINAL_PATH]) else: input_stack_path = row[star.UCSF.IMAGE_ORIGINAL_PATH] with mrc.ZSliceReader(input_stack_path) as reader: i = row[star.UCSF.IMAGE_ORIGINAL_INDEX] writer.write(reader.read(i)) elif fn.endswith(".par"): if args.stack_path is None: log.error(".par file input requires --stack-path") return 1 df = metadata.par2star(metadata.parse_fx_par(fn), data_path=args.stack_path) # star.set_original_fields(df, inplace=True) # Redundant. star.augment_star_ucsf(df) elif fn.endswith(".csv"): return 1 elif fn.endswith(".cs"): return 1 else: if fn.endswith(".mrcs"): with mrc.ZSliceReader(fn) as reader: for img in reader: writer.write(img) df = pd.DataFrame( {star.UCSF.IMAGE_ORIGINAL_INDEX: np.arange(reader.nz)}) df[star.UCSF.IMAGE_ORIGINAL_PATH] = fn else: print("Unrecognized input file type") return 1 if args.star is not None: df[star.UCSF.IMAGE_INDEX] = np.arange(first_ptcl, first_ptcl + df.shape[0]) if args.abs_path: df[star.UCSF.IMAGE_PATH] = writer.path else: df[star.UCSF.IMAGE_PATH] = os.path.relpath(writer.path, os.path.dirname(args.star)) df["index"] = df[star.UCSF.IMAGE_INDEX] star.simplify_star_ucsf(df) dfs.append(df) first_ptcl += df.shape[0] if args.star is not None: df = pd.concat(dfs, join="inner") # df = pd.concat(dfs) # df = df.dropna(df, axis=1, how="any") if not args.relion2: # Relion 3.1 style output. df = star.remove_deprecated_relion2(df, inplace=True) star.write_star(args.star, df, resort_records=False, optics=True) else: df = star.remove_new_relion31(df, inplace=True) star.write_star(args.star, df, resort_records=False, optics=False) return 0
def main(args): """ Projection subtraction program entry point. :param args: Command-line arguments parsed by ArgumentParser.parse_args() :return: Exit status """ log = logging.getLogger('root') hdlr = logging.StreamHandler(sys.stdout) log.addHandler(hdlr) log.setLevel(logging.getLevelName(args.loglevel.upper())) if args.dest is None and args.suffix == "": args.dest = "" args.suffix = "_subtracted" log.info("Reading particle .star file") df = star.parse_star(args.input, keep_index=False) star.augment_star_ucsf(df) if not args.original: df[star.UCSF.IMAGE_ORIGINAL_PATH] = df[star.UCSF.IMAGE_PATH] df[star.UCSF.IMAGE_ORIGINAL_INDEX] = df[star.UCSF.IMAGE_INDEX] df.sort_values(star.UCSF.IMAGE_ORIGINAL_PATH, inplace=True, kind="mergesort") gb = df.groupby(star.UCSF.IMAGE_ORIGINAL_PATH) df[star.UCSF.IMAGE_INDEX] = gb.cumcount() df[star.UCSF.IMAGE_PATH] = df[star.UCSF.IMAGE_ORIGINAL_PATH].map( lambda x: os.path.join( args.dest, args.prefix + os.path.basename(x).replace( ".mrcs", args.suffix + ".mrcs"))) if args.submap_ft is None: log.info("Reading volume") submap = mrc.read(args.submap, inc_header=False, compat="relion") if args.submask is not None: log.info("Masking volume") submask = mrc.read(args.submask, inc_header=False, compat="relion") submap *= submask log.info("Preparing 3D FFT of volume") submap_ft = vop.vol_ft(submap, pfac=args.pfac, threads=min(args.threads, cpu_count())) log.info("Finished 3D FFT of volume") else: log.info("Loading 3D FFT from %s" % args.submap_ft) submap_ft = np.load(args.submap_ft) log.info("Loaded 3D FFT from %s" % args.submap_ft) sz = (submap_ft.shape[0] - 3) // args.pfac maxshift = np.round(np.max(np.abs(df[star.Relion.ORIGINS].values))) if args.crop is not None and sz < 2 * maxshift + args.crop: log.error("Some shifts are too large to crop (maximum crop is %d)" % (sz - 2 * maxshift)) return 1 sx, sy = np.meshgrid(np.fft.rfftfreq(sz), np.fft.fftfreq(sz)) s = np.sqrt(sx**2 + sy**2) r = s * sz r = np.round(r).astype(np.int64) r[r > sz // 2] = sz // 2 + 1 nr = np.max(r) + 1 a = np.arctan2(sy, sx) if args.refmap is not None: coefs_method = 1 if args.refmap_ft is None: refmap = mrc.read(args.refmap, inc_header=False, compat="relion") refmap_ft = vop.vol_ft(refmap, pfac=args.pfac, threads=min(args.threads, cpu_count())) else: log.info("Loading 3D FFT from %s" % args.refmap_ft) refmap_ft = np.load(args.refmap_ft) log.info("Loaded 3D FFT from %s" % args.refmap_ft) else: coefs_method = 0 refmap_ft = np.empty(submap_ft.shape, dtype=submap_ft.dtype) apix = star.calculate_apix(df) log.info("Computed pixel size is %f A" % apix) log.debug("Grouping particles by output stack") gb = df.groupby(star.UCSF.IMAGE_PATH) iothreads = threading.BoundedSemaphore(args.io_thread_pairs) qsize = args.io_queue_length fftthreads = args.fft_threads def init(): global tls tls = threading.local() log.info("Instantiating thread pool with %d workers" % args.threads) pool = Pool(processes=args.threads, initializer=init) threads = [] log.info("Performing projection subtraction") try: for fname, particles in gb: log.debug("Instantiating queue") queue = Queue.Queue(maxsize=qsize) log.debug("Create producer for %s" % fname) prod = threading.Thread(target=producer, args=(pool, queue, submap_ft, refmap_ft, fname, particles, sx, sy, s, a, apix, coefs_method, r, nr, fftthreads, args.crop, args.pfac)) log.debug("Create consumer for %s" % fname) cons = threading.Thread(target=consumer, args=(queue, fname, apix, iothreads)) threads.append((prod, cons)) iothreads.acquire() log.debug("iotheads at %d" % iothreads._Semaphore__value) log.debug("Start consumer for %s" % fname) cons.start() log.debug("Start producer for %s" % fname) prod.start() except KeyboardInterrupt: log.debug("Main thread wants out!") for pair in threads: for thread in pair: try: thread.join() except RuntimeError as e: log.debug(e) pool.close() pool.join() pool.terminate() log.info("Finished projection subtraction") log.info("Writing output .star file") if args.crop is not None: df = star.recenter(df, inplace=True) star.simplify_star_ucsf(df) star.write_star(args.output, df) return 0
def main(args): """ Projection subtraction program entry point. :param args: Command-line arguments parsed by ArgumentParser.parse_args() :return: Exit status """ log = logging.getLogger('root') hdlr = logging.StreamHandler(sys.stdout) log.addHandler(hdlr) log.setLevel(logging.getLevelName(args.loglevel.upper())) log.debug("Reading particle .star file") df = parse_star(args.input, keep_index=False) df.reset_index(inplace=True) df["rlnImageOriginalName"] = df["rlnImageName"] df["ucsfOriginalParticleIndex"], df["ucsfOriginalImagePath"] = \ df["rlnImageOriginalName"].str.split("@").str df["ucsfOriginalParticleIndex"] = pd.to_numeric( df["ucsfOriginalParticleIndex"]) df.sort_values("rlnImageOriginalName", inplace=True, kind="mergesort") gb = df.groupby("ucsfOriginalImagePath") df["ucsfParticleIndex"] = gb.cumcount() + 1 df["ucsfImagePath"] = df["ucsfOriginalImagePath"].map( lambda x: os.path.join( args.dest, args.prefix + os.path.basename(x).replace( ".mrcs", args.suffix + ".mrcs"))) df["rlnImageName"] = df["ucsfParticleIndex"].map( lambda x: "%.6d" % x).str.cat(df["ucsfImagePath"], sep="@") log.debug("Read particle .star file") if args.submap_ft is None: submap = mrc.read(args.submap, inc_header=False, compat="relion") submap_ft = vol_ft(submap, threads=min(args.threads, cpu_count())) else: log.debug("Loading %s" % args.submap_ft) submap_ft = np.load(args.submap_ft) log.debug("Loaded %s" % args.submap_ft) sz = submap_ft.shape[0] // 2 - 1 sx, sy = np.meshgrid(np.fft.rfftfreq(sz), np.fft.fftfreq(sz)) s = np.sqrt(sx**2 + sy**2) r = s * sz r = np.round(r).astype(np.int64) r[r > sz // 2] = sz // 2 + 1 nr = np.max(r) + 1 a = np.arctan2(sy, sx) if args.refmap is not None: coefs_method = 1 if args.refmap_ft is None: refmap = mrc.read(args.refmap, inc_header=False, compat="relion") refmap_ft = vol_ft(refmap, threads=min(args.threads, cpu_count())) else: log.debug("Loading %s" % args.refmap_ft) refmap_ft = np.load(args.refmap_ft) log.debug("Loaded %s" % args.refmap_ft) else: coefs_method = 0 refmap_ft = np.empty(submap_ft.shape, dtype=submap_ft.dtype) apix = calculate_apix(df) log.debug("Constructing particle metadata references") # npart = df.shape[0] idx = df["ucsfOriginalParticleIndex"].values stack = df["ucsfOriginalImagePath"].values.astype(np.str, copy=False) def1 = df["rlnDefocusU"].values def2 = df["rlnDefocusV"].values angast = df["rlnDefocusAngle"].values phase = df["rlnPhaseShift"].values kv = df["rlnVoltage"].values ac = df["rlnAmplitudeContrast"].values cs = df["rlnSphericalAberration"].values az = df["rlnAngleRot"].values el = df["rlnAngleTilt"].values sk = df["rlnAnglePsi"].values xshift = df["rlnOriginX"].values yshift = df["rlnOriginY"].values new_idx = df["ucsfParticleIndex"].values new_stack = df["ucsfImagePath"].values.astype(np.str, copy=False) log.debug("Grouping particles by output stack") gb = df.groupby("ucsfImagePath") iothreads = threading.BoundedSemaphore(args.io_thread_pairs) qsize = args.io_queue_length fftthreads = args.fft_threads # pyfftw.interfaces.cache.enable() log.debug("Instantiating worker pool") pool = Pool(processes=args.threads) threads = [] try: for fname, particles in gb.indices.iteritems(): log.debug("Instantiating queue") queue = Queue.Queue(maxsize=qsize) log.debug("Create producer for %s" % fname) prod = threading.Thread( target=producer, args=(pool, queue, submap_ft, refmap_ft, fname, particles, idx, stack, sx, sy, s, a, apix, def1, def2, angast, phase, kv, ac, cs, az, el, sk, xshift, yshift, new_idx, new_stack, coefs_method, r, nr, fftthreads)) log.debug("Create consumer for %s" % fname) cons = threading.Thread(target=consumer, args=(queue, fname, apix, fftthreads, iothreads)) threads.append((prod, cons)) iothreads.acquire() log.debug("iotheads at %d" % iothreads._Semaphore__value) log.debug("Start consumer for %s" % fname) cons.start() log.debug("Start producer for %s" % fname) prod.start() except KeyboardInterrupt: log.debug("Main thread wants out!") for pair in threads: for thread in pair: try: thread.join() except RuntimeError as e: log.debug(e) pool.close() pool.join() pool.terminate() df.drop([c for c in df.columns if "ucsf" in c or "eman" in c], axis=1, inplace=True) df.set_index("index", inplace=True) df.sort_index(inplace=True, kind="mergesort") write_star(args.output, df, reindex=True) return 0
def main(args): log = logging.getLogger(__name__) log.setLevel(logging.INFO) hdlr = logging.StreamHandler(sys.stdout) if args.quiet: hdlr.setLevel(logging.WARNING) else: hdlr.setLevel(logging.INFO) log.addHandler(hdlr) if args.markers is None and args.target is None and args.sym is None: log.error( "A marker or symmetry group must be provided via --target, --markers, or --sym" ) return 1 elif args.sym is None and args.markers is None and args.boxsize is None and args.origin is None: log.error( "An origin must be provided via --boxsize, --origin, or --markers") return 1 elif args.sym is not None and args.markers is None and args.target is None and \ (args.boxsize is not None or args.origin is not None): log.warn("Symmetry expansion alone will ignore --target or --origin") if args.target is not None: try: args.target = np.array( [np.double(tok) for tok in args.target.split(",")]) except: log.error( "Target must be comma-separated list of x,y,z coordinates") return 1 if args.origin is not None: if args.boxsize is not None: logger.warn("--origin supersedes --boxsize") try: args.origin = np.array( [np.double(tok) for tok in args.origin.split(",")]) except: log.error( "Origin must be comma-separated list of x,y,z coordinates") return 1 if args.marker_sym is not None: args.marker_sym = relion_symmetry_group(args.marker_sym) star = parse_star(args.input, keep_index=False) if args.apix is None: args.apix = calculate_apix(star) if args.apix is None: logger.warn( "Could not compute pixel size, default is 1.0 Angstroms per pixel" ) args.apix = 1.0 if args.cls is not None: star = select_classes(star, args.cls) cmms = [] if args.markers is not None: cmmfiles = glob.glob(args.markers) for cmmfile in cmmfiles: for cmm in parse_cmm(cmmfile): cmms.append(cmm / args.apix) if args.target is not None: cmms.append(args.target / args.apix) stars = [] if len(cmms) > 0: if args.origin is not None: args.origin /= args.apix elif args.boxsize is not None: args.origin = np.ones(3) * args.boxsize / 2 else: log.warn("Using first marker as origin") if len(cmms) == 1: log.error( "Using first marker as origin, expected at least two markers" ) return 1 args.origin = cmms[0] cmms = cmms[1:] markers = [cmm - args.origin for cmm in cmms] if args.marker_sym is not None and len(markers) == 1: markers = [op.dot(markers[0]) for op in args.marker_sym] elif args.marker_sym is not None: log.error( "Exactly one marker is required for symmetry-derived subparticles" ) return 1 rots = [euler2rot(*np.deg2rad(r[1])) for r in star[ANGLES].iterrows()] #origins = star[ORIGINS].copy() for m in markers: d = np.linalg.norm(m) ax = m / d op = euler2rot( *np.array([np.arctan2(ax[1], ax[0]), np.arccos(ax[2]), 0.])) stars.append(transform_star(star, op.T, -d, rots=rots)) if args.sym is not None: args.sym = relion_symmetry_group(args.sym) if len(stars) > 0: stars = [ se for se in subparticle_expansion( s, args.sym, -args.displacement / args.apix) for s in stars ] else: stars = list( subparticle_expansion(star, args.sym, -args.displacement / args.apix)) if args.recenter: for s in stars: recenter(s, inplace=True) if args.suffix is None and not args.skip_join: if len(stars) > 1: star = interleave(stars) else: star = stars[0] write_star(args.output, star) else: for i, star in enumerate(stars): write_star(os.path.join(args.output, args.suffix + "_%d" % i), star) return 0
def main(args): meta = parse_metadata(args.input) # Read cryosparc metadata file. meta["data_input_idx"] = [ "%.6d" % (i + 1) for i in meta["data_input_idx"] ] # Reformat particle idx for Relion. if "data_input_relpath" not in meta.columns: if args.data_path is None: print( "Data path missing, use --data-path to specify particle stack path" ) return 1 meta["data_input_relpath"] = args.data_path meta["data_input_relpath"] = meta["data_input_idx"].str.cat( meta["data_input_relpath"], sep="@") # Construct _rlnImageName field. # Take care of trivial mappings. rlnheaders = [ general[h] for h in meta.columns if h in general and general[h] is not None ] star = meta[[ h for h in meta.columns if h in general and general[h] is not None ]].copy() star.columns = rlnheaders if "rlnRandomSubset" in star.columns: star["rlnRandomSubset"] = star["rlnRandomSubset"].apply( lambda x: ord(x) - 64) if "rlnPhaseShift" in star.columns: star["rlnPhaseShift"] = np.rad2deg(star["rlnPhaseShift"]) # general class assignments and other model parameters. phic = meta[[h for h in meta.columns if "phiC" in h ]] # Posterior probability over class assignments. if len(phic.columns) > 0: # Check class assignments exist in input. # phic.columns = [int(h[21]) for h in meta.columns if "phiC" in h] phic.columns = range(len(phic.columns)) cls = phic.idxmax(axis=1) for p in model: if model[p] is not None: pspec = p.split("model")[1] param = meta[[h for h in meta.columns if pspec in h]] if len(param.columns) > 0: param.columns = phic.columns star[model[p]] = param.lookup(param.index, cls) star[ "rlnClassNumber"] = cls + 1 # Compute most probable classes and add one for Relion indexing. else: for p in model: if model[p] is not None and p in meta.columns: star[model[p]] = meta[p] star["rlnClassNumber"] = 1 if args.cls is not None: star = select_classes(star, args.cls) # Convert axis-angle representation to Euler angles (degrees). if star.columns.intersection(angles).size == len(angles): star[angles] = np.rad2deg(star[angles].apply( lambda x: rot2euler(expmap(x)), axis=1, raw=True, broadcast=True)) if args.minphic is not None: mask = np.all(phic < args.minphic, axis=1) if args.drop_bad: star.drop(star[mask].index, inplace=True) # Delete low-confidence particles. else: star.loc[ mask, "rlnClassNumber"] = 0 # Set low-confidence particles to dummy class. if args.transform is not None: r = np.array(json.loads(args.transform)) star = transform_star(star, r, inplace=True) # Write Relion .star file with correct headers. write_star(args.output, star, reindex=True) return 0
def main(args): meta = parse_metadata(args.input) # Read cryosparc metadata file. meta["data_input_idx"] = [ "%.6d" % (i + 1) for i in meta["data_input_idx"] ] # Reformat particle idx for Relion. if "data_input_relpath" not in meta.columns: if args.data_path is None: print( "Data path missing, use --data-path to specify particle stack path" ) return 1 meta["data_input_relpath"] = args.data_path meta["data_input_relpath"] = meta["data_input_idx"].str.cat( meta["data_input_relpath"], sep="@") # Construct rlnImageName field. # Take care of trivial mappings. rlnheaders = [ general[h] for h in meta.columns if h in general and general[h] is not None ] df = meta[[ h for h in meta.columns if h in general and general[h] is not None ]].copy() df.columns = rlnheaders if "rlnRandomSubset" in df.columns: df["rlnRandomSubset"] = df["rlnRandomSubset"].apply( lambda x: ord(x) - 64) if "rlnPhaseShift" in df.columns: df["rlnPhaseShift"] = np.rad2deg(df["rlnPhaseShift"]) # Class assignments and other model parameters. phic = meta[[h for h in meta.columns if "phiC" in h ]] # Posterior probability over class assignments. if len(phic.columns) > 0: # Check class assignments exist in input. # phic.columns = [int(h[21]) for h in meta.columns if "phiC" in h] phic.columns = range(len(phic.columns)) cls = phic.idxmax(axis=1) for p in model: if model[p] is not None: pspec = p.split("model")[1] param = meta[[h for h in meta.columns if pspec in h]] if len(param.columns) > 0: param.columns = phic.columns df[model[p]] = param.lookup(param.index, cls) df["rlnClassNumber"] = cls + 1 # Add one for Relion indexing. else: for p in model: if model[p] is not None and p in meta.columns: df[model[p]] = meta[p] df["rlnClassNumber"] = 1 if args.cls is not None: df = star.select_classes(df, args.cls) # Convert axis-angle representation to Euler angles (degrees). if df.columns.intersection(star.Relion.ANGLES).size == len( star.Relion.ANGLES): df[star.Relion.ANGLES] = np.rad2deg(df[star.Relion.ANGLES].apply( lambda x: rot2euler(expmap(x)), axis=1, raw=True, broadcast=True)) if args.minphic is not None: mask = np.all(phic < args.minphic, axis=1) if args.keep_bad: df.loc[mask, "rlnClassNumber"] = 0 else: df.drop(df[mask].index, inplace=True) if args.copy_micrograph_coordinates is not None: coord_star = pd.concat( (star.parse_star(inp, keep_index=False) for inp in glob(args.copy_micrograph_coordinates)), join="inner") df = star.smart_merge(df, coord_star, fields=star.Relion.MICROGRAPH_COORDS) if args.transform is not None: r = np.array(json.loads(args.transform)) df = star.transform_star(df, r, inplace=True) # Write Relion .star file with correct headers. star.write_star(args.output, df, reindex=True) return 0