Examples with ContainsLProteinChain edu.sdsc.mmtf.spark.filters.ContainsLProteinChain used on opensource projects

Example 6 with ContainsLProteinChain

use of edu.sdsc.mmtf.spark.filters.ContainsLProteinChain in project mmtf-spark by sbl-sdsc.

the class WaterInteractions method main.

public static void main(String[] args) throws IOException, ParseException {
    String timeStamp = new SimpleDateFormat("yyyyMMdd_HHmm").format(Calendar.getInstance().getTime());
    long start = System.nanoTime();
    // process command line options (defaults are provided)
    CommandLine cmd = getCommandLine(args);
    String outputPath = cmd.getOptionValue("output-path");
    System.out.println(outputPath);
    String resolution = cmd.getOptionValue("resolution", "2");
    String minInteractions = cmd.getOptionValue("min-interactions", "2");
    String maxInteractions = cmd.getOptionValue("max-interactions", "4");
    String distanceCutoff = cmd.getOptionValue("distance-cutoff", "3");
    String bFactorCutoff = cmd.getOptionValue("b-factor-cutoff", "1.645");
    boolean includeWaters = cmd.hasOption("include-waters");
    // get path to MMTF Hadoop Sequence file
    String path = MmtfReader.getMmtfFullPath();
    // initialize Spark
    SparkConf conf = new SparkConf().setMaster("local[*]").setAppName(WaterInteractions.class.getSimpleName());
    JavaSparkContext sc = new JavaSparkContext(conf);
    // read PDB structures and filter by resolution and only include proteins
    JavaPairRDD<String, StructureDataInterface> pdb = MmtfReader.readSequenceFile(path, sc).filter(new Resolution(0.0, Float.parseFloat(resolution))).filter(new ContainsLProteinChain(true));
    // setup interaction criteria
    InteractionFilter filter = new InteractionFilter();
    filter.setDistanceCutoff(Float.parseFloat(distanceCutoff));
    filter.setNormalizedbFactorCutoff(Float.parseFloat(bFactorCutoff));
    filter.setMinInteractions(Integer.parseInt(minInteractions));
    filter.setMaxInteractions(Integer.parseInt(maxInteractions));
    filter.setQueryGroups(true, "HOH");
    // only use water oxygen
    filter.setQueryElements(true, "O");
    filter.setTargetElements(true, "O", "N", "S");
    // exclude "uninteresting" ligands
    Set<String> prohibitedGroups = new HashSet<>();
    prohibitedGroups.addAll(ExcludedLigandSets.ALL_GROUPS);
    if (!includeWaters) {
        prohibitedGroups.add("HOH");
    }
    filter.setProhibitedTargetGroups(prohibitedGroups);
    // calculate interactions
    Dataset<Row> data = GroupInteractionExtractor.getInteractions(pdb, filter);
    // keep only interactions with at least one organic ligand and one protein interaction
    data = filterBridgingWaterInteractions(data, maxInteractions).cache();
    // show some results
    data.show(50);
    System.out.println("Hits(all): " + data.count());
    // save interactions to a .parquet file
    String waterTag = includeWaters ? "_w" : "";
    String filename = outputPath + "/water_pl" + "_r" + resolution + "_d" + distanceCutoff + "_b" + bFactorCutoff + "_i" + minInteractions + maxInteractions + waterTag + "_" + timeStamp + ".parquet";
    System.out.println("Saving results to: " + filename);
    data.coalesce(1).write().mode("overwrite").format("parquet").save(filename);
    // exit Spark
    sc.close();
    long end = System.nanoTime();
    System.out.println("Time: " + TimeUnit.NANOSECONDS.toSeconds(end - start) + " sec.");
}

Example 7 with ContainsLProteinChain

use of edu.sdsc.mmtf.spark.filters.ContainsLProteinChain in project mmtf-spark by sbl-sdsc.

the class SecondaryStructureElementDemo method main.

public static void main(String[] args) throws IOException {
    long start = System.nanoTime();
    SparkConf conf = new SparkConf().setMaster("local[1]").setAppName(CustomReportDemo.class.getSimpleName());
    JavaSparkContext sc = new JavaSparkContext(conf);
    // single protein chain
    List<String> pdbIds = Arrays.asList("1STP");
    JavaPairRDD<String, StructureDataInterface> pdb = MmtfReader.downloadReducedMmtfFiles(pdbIds, sc).cache();
    pdb = pdb.flatMapToPair(new StructureToPolymerChains()).filter(new ContainsLProteinChain());
    Dataset<Row> ds = SecondaryStructureElementExtractor.getDataset(pdb, "E", 6);
    // show the top 50 rows of this dataset
    ds.show(50, false);
    long end = System.nanoTime();
    System.out.println("Time: " + TimeUnit.NANOSECONDS.toSeconds(end - start) + " sec.");
    sc.close();
}

Example 8 with ContainsLProteinChain

use of edu.sdsc.mmtf.spark.filters.ContainsLProteinChain in project mmtf-spark by sbl-sdsc.

the class SecondaryStructureSegmentDemo method main.

public static void main(String[] args) throws IOException {
    long start = System.nanoTime();
    SparkConf conf = new SparkConf().setMaster("local[*]").setAppName(CustomReportDemo.class.getSimpleName());
    JavaSparkContext sc = new JavaSparkContext(conf);
    // single protein chain
    List<String> pdbIds = Arrays.asList("1STP");
    JavaPairRDD<String, StructureDataInterface> pdb = MmtfReader.downloadReducedMmtfFiles(pdbIds, sc).cache();
    pdb = pdb.flatMapToPair(new StructureToPolymerChains()).filter(new ContainsLProteinChain());
    int segmentLength = 25;
    Dataset<Row> ds = SecondaryStructureSegmentExtractor.getDataset(pdb, segmentLength);
    // show the top 50 rows of this dataset
    ds.show(50, false);
    long end = System.nanoTime();
    System.out.println("Time: " + TimeUnit.NANOSECONDS.toSeconds(end - start) + " sec.");
    sc.close();
}

Example 9 with ContainsLProteinChain

use of edu.sdsc.mmtf.spark.filters.ContainsLProteinChain in project mm-dev by sbl-sdsc.

the class ArgLigandInteractions method main.

public static void main(String[] args) throws IOException, ParseException {
    String timeStamp = new SimpleDateFormat("yyyyMMdd_HHmm").format(Calendar.getInstance().getTime());
    long start = System.nanoTime();
    // process command line options (defaults are provided)
    CommandLine cmd = getCommandLine(args);
    String outputPath = cmd.getOptionValue("output-path");
    System.out.println(outputPath);
    String resolution = cmd.getOptionValue("resolution", "2");
    String minInteractions = cmd.getOptionValue("min-interactions", "2");
    String maxInteractions = cmd.getOptionValue("max-interactions", "4");
    String distanceCutoff = cmd.getOptionValue("distance-cutoff", "3");
    String bFactorCutoff = cmd.getOptionValue("b-factor-cutoff", "1.645");
    boolean includeWaters = cmd.hasOption("include-waters");
    // get path to MMTF Hadoop Sequence file
    String path = MmtfReader.getMmtfFullPath();
    // initialize Spark
    SparkConf conf = new SparkConf().setMaster("local[*]").setAppName(ArgLigandInteractions.class.getSimpleName());
    JavaSparkContext sc = new JavaSparkContext(conf);
    // read PDB structures and filter by resolution and only include proteins
    JavaPairRDD<String, StructureDataInterface> pdb = MmtfReader.readSequenceFile(path, sc).filter(new Resolution(0.0, Float.parseFloat(resolution))).filter(new ContainsLProteinChain(true));
    // setup interaction criteria
    InteractionFilter filter = new InteractionFilter();
    filter.setDistanceCutoff(Float.parseFloat(distanceCutoff));
    filter.setNormalizedbFactorCutoff(Float.parseFloat(bFactorCutoff));
    filter.setMinInteractions(Integer.parseInt(minInteractions));
    filter.setMaxInteractions(Integer.parseInt(maxInteractions));
    filter.setQueryGroups(true, "ARG");
    // only use water oxygen
    filter.setQueryElements(true, "N");
    filter.setTargetElements(true, "O", "N", "S");
    filter.setTargetGroups(false, new HashSet<>(PolymerComposition.AMINO_ACIDS_20));
    // exclude "uninteresting" ligands
    Set<String> prohibitedGroups = new HashSet<>();
    prohibitedGroups.addAll(ExcludedLigandSets.ALL_GROUPS);
    if (!includeWaters) {
        prohibitedGroups.add("HOH");
    }
    filter.setProhibitedTargetGroups(prohibitedGroups);
    // calculate interactions
    Dataset<Row> data = GroupInteractionExtractor.getInteractions(pdb, filter);
    // only consider interactions with ARG sidechain nitrogens
    data = data.filter("atom0 = 'NE' OR atom0 = 'NH1' OR atom0 = 'NH2'");
    // the interacting group should be an organic ligand (LGO)
    data = data.filter("type1 = 'LGO'");
    data = data.select("pdbId", "atom0", "groupNum0", "chain0", "atom1", "group1", "groupNum1", "chain1", "distance1");
    // data.show(50);
    Dataset<Row> data2 = data;
    Dataset<Row> joint = data.join(data2, (data.col("pdbId").equalTo(data2.col("pdbId"))).and(data.col("atom0").notEqual(data2.col("atom0"))).and(data.col("groupNum1").equalTo(data2.col("groupNum1")).and(data.col("chain1").equalTo(data2.col("chain1")).and(data.col("atom1").notEqual(data2.col("atom1"))))));
    joint.show(100);
    // data = data.select("pdbId",
    // "atom0", "groupNum0", "chain0",
    // "atom1", "groupNum1", "chain1", "distance1",
    // "atom2", "groupNum2", "chain2", "distance2");
    // 
    // // only consider interactions with ARG sidechain nitrogens
    // data = data.filter("atom0 = 'NE' OR atom0 = 'NH1' OR atom0 = 'NH2'");
    // 
    // // the interacting group should be an organic ligand (LGO)
    // data = data.filter("type1 = 'LGO' AND type2 = 'LGO'").cache();
    // 
    // // the two interacting atoms must come from the same group and chain
    // data = data.filter("group1 = group2 AND groupNum1 = groupNum2 AND chain1 = chain2");
    // Dataset<Row> data2 = data;
    // Dataset<Row> joint = data.join(data2,
    // data.col("pdbId").equalTo(data2.col("pdbId")).and
    // (data.col("groupNum1").equalTo(data2.col("groupNum1")).and
    // (data.col("chain1").equalTo(data2.col("chain1")))
    // ));
    // joint.show(100);
    // RelationalGroupedDataset groupBy = data.groupBy("pdbId", "groupNum0", "chain0", "group1", "groupNum1");
    // groupBy.count().show(1000);
    // show some results
    // data.show(50);
    // System.out.println("Hits(all): " + data.count());
    // 
    // // save interactions to a .parquet file
    // String waterTag = includeWaters ? "_w" : "";
    // String filename = outputPath + "/arg_lig" + "_r" + resolution
    // + "_d" + distanceCutoff
    // + "_b" + bFactorCutoff + "_i" + minInteractions + maxInteractions + waterTag + "_" + timeStamp + ".parquet";
    // System.out.println("Saving results to: " + filename);
    // data.coalesce(1).write().mode("overwrite").format("parquet").save(filename);
    // exit Spark
    sc.close();
    long end = System.nanoTime();
    System.out.println("Time: " + TimeUnit.NANOSECONDS.toSeconds(end - start) + " sec.");
}