fix pdbe's updated cifs parsing problem using forked biojava with new parseSites parsing param

This commit is contained in:
rdk
2025-02-23 16:01:36 +01:00
parent 820239d2b2
commit 86eb8bdae1
4 changed files with 8820 additions and 4 deletions

View File

@@ -22,7 +22,11 @@ apply plugin: "com.github.ben-manes.versions" // ./gradlew dependencyUpdates
group = 'cz.siret'
<<<<<<< HEAD
version = '2.5.1-rc.1'
=======
version = '2.5.1-dev.3-parsFix'
>>>>>>> 2e40ef0e (fix pdbe's updated cifs parsing problem using forked biojava with new parseSites parsing param)
description = 'Ligand binding site prediction based on machine learning.'

View File

@@ -28,13 +28,15 @@ class PdbUtils {
static final FileParsingParameters PARSING_PARAMS = new FileParsingParameters()
static {
PARSING_PARAMS.setAlignSeqRes(false); // should the ATOM and SEQRES residues be aligned when creating the internal data model?
PARSING_PARAMS.setParseSecStruc(false); // should secondary structure getByID parsed from the file
PARSING_PARAMS.setAlignSeqRes(false) // should the ATOM and SEQRES residues be aligned when creating the internal data model?
PARSING_PARAMS.setParseSecStruc(false) // should secondary structure getByID parsed from the file
PARSING_PARAMS.setCreateAtomBonds(false)
PARSING_PARAMS.setCreateAtomCharges(false)
PARSING_PARAMS.setParseBioAssembly(false)
PARSING_PARAMS.parseSites = false // using custom forked build with new parseSites param
disableBiojavaFetching()
//PARSING_PARAMS.setLoadChemCompInfo(Params.inst.biojava_load_chem_info); // info about modified amino acid residues, not available in BioJava4
}

View File

@@ -1,5 +1,6 @@
package cz.siret.prank.domain
import cz.siret.prank.utils.PdbUtils
import groovy.transform.CompileStatic
import groovy.util.logging.Slf4j
import org.junit.jupiter.api.Disabled
@@ -22,7 +23,10 @@ class ProteinParsingTest {
void testParsePdbeUpdatedCif() {
Protein protein = Protein.load("$DIR/4gqq.cif")
Protein protein2 = Protein.load("$DIR/4gqq_updated.cif") // fails in 2.5, still fails with BioJava 7.1.4 and 7.2.0
PdbUtils.PARSING_PARAMS.parseSites = false // using custom forked build with new parseSites param
//Protein protein2 = Protein.load("$DIR/4gqq_updated.cif") // fails in 2.5, still fails with BioJava 7.1.4 and 7.2.2
Protein protein2 = Protein.load("$DIR/4gqq_updated_2025.cif") // fails in 2.5, still fails with BioJava 7.1.4 and 7.2.2
}
@Test
@@ -35,7 +39,8 @@ class ProteinParsingTest {
/**
* https://github.com/rdk/p2rank/issues/77
*
* 2.5 supposedly fails with org.rcsb.cif.EmptyColumnException: column pdbx_PDB_id_code is undefined
* P2Rank 2.5 (using BioJava 7.1.3) fails with org.rcsb.cif.EmptyColumnException: column pdbx_PDB_id_code is undefined
* Already fixed in BioJava 7.2.2
*/
@Test
void testParse8zz7Cif() {

File diff suppressed because it is too large Load Diff