Bioinformatics Group - Thesis projects

1 Bioinformatics Group - Thesis projectsLast updated: April 30th, 2021 Reconstructing symbiont strains from long-read metagenomicsAnnotation and analysis of the blue shark (Prionaceglauca) genomeThe evolution of auxiliary metabolic genes in bacteriophagesBiG-SCAPE updates to a successful software forlarge-scale genome miningGene content variation mediated by recombination inbacteriophage evolutionInferring viral recombination from metagenomesMachine learning to predict enzyme regioselectivityStrategic genome mining for novel antimicrobial compoundsMining microbiomes for novel peptidic natural productsA novel metabolite annotation approach combining LC-MSand LC-MS/MS dataLatent space models to link metabolite structuresand MS spectraMS2 ChemClass: automated mass spectral-based chemicalclass annotationElemental Formula and Mass Difference-enhanced SubstructureDiscovery in MetabolomicsProfilesTracing Drug and Food Substructures & their Biotransformationsin UrineToward Natural Products aware Chemical FingerprintsLocal assembly of QTL regions in plantsUsing machine learning to predict underlying factorsfor plant meiotic recombination ratevariationExplaining functional properties of non-domain proteinsequence regionsMatrix factorization for gene function predictionThe origins of promiscuity in fragrance-producingenzymesProtein-protei n interaction prediction using co-evolutionA phylogenetic framework for linking genes to moleculesin large-scale genomic/metabolomicdatasetsPangenomic QTL analysisIntegrative QTL analysisNovel enzymes for fragrance and flavourConnecting

2 Transcription factor - DNA interactionto flowering time regulationFinding genes related to regenerationPrediction in pangenome graphsThe INSIDE study: can probiotics improve quality oflife of ileostoma patients?High throughput single cell RNA sequencing to studyplant root developmentTiming of transcriptional regulation during earlyroot nodule organogenesisProtein interactions involved in SOBIR1/BAK1-mediatedplant immunityUsing protein structures and interactions to screenfor mechanosensorsImproving deep learning based genomic prediction modelsby incorporating prior knowledgePlant-specific basecalling of Oxford Nanopore dataPangenomics in tetraploidsSequencing-based high density marker developmentFold2vec: protein structure embedding for deep learningDeep learning on protein-DNA sequence pairs usingthe encoder-decoderDeep learning with DNABERT to predict genomic propertiesReconstructing symbiont strains from long-read metagenomicsSupervisorAnne KupczokTypeData analysis, MethodologyRequirementsAdvanced BioinformaticsSkillsGenome assembly, Metagenomics, ComparativegenomicsTimestampMarch 2021 DescriptionInthedarkpartofthedeepocean, ,Bathymodiolusmusselsharborchemosyntheti csulfur-oxidizing(SOX)andmethane-oxidizi ng(MOX)bacteriainaspecialorgan, ,metagenomicanalysesofmultipleBathymodio lusspeciesshowedthatdifferentSOXandMOXst rainscanbepresentwithinanindividualmusse l( ; ).

3 Strainsaredifferentvariantsofaspeciesand mightharborimportantfunctionaldifference s( ).Notably,co-occurringSOXstrainscandiffe rinthecontentofgenesinvolvedinenergyandn utrientutilizationandviraldefencemechani sms( ).SOXgenomeswerealsofoundtocontainhighnu mbersofmobilegeneticelementssuchastransp osases,integrases,restriction-modificati onsystems,andtoxin-relatedgenes,wherethe latter are also linked to mobile genetic elements(Sayavedra et al. 2015).Thepresenceofdifferentstrainsandof mobilegeneticelementsresultedinhighlyfra gmentedassembliesforshort-readmetagenome s, ( ) ,MaxPlanckInstituteforMarine Microbiology Bremen, R, Romano S, Sayavedra L, Porras M G, KupczokA, Tegetmeyer HE, Dubilier N, Petersen J. diversity enables multiple symbiont strainsto coexist in deep-sea mussels. Nature Microbiology4:2487 M, Bickhart DM, Behsaz B, Gurevich A, RaykoM, Shin SB, Kuhn K, Yuan J, Polevikov E, Smith TPL,etal.

4 2020. metaFlye: scalable long-read metagenomeassembly using repeat graphs. Nature Methods 17: 1103 Picazo D, Dagan T, Ansorge R, Petersen JM,Dubilier N, Kupczok A. 2019. Horizontally transmittedsymbiontpopulations in deep-sea mussels are genetically J 13:2954 TV, Ferretti P, Maistrenko OM, Bork P. within species: interpreting strains Reviews Microbiology 18:491 L, Kleiner M, Ponnudurai R, Wetzel S, PelletierE, Barbe V, Satoh N, Shoguchi E, Fink D, BreusingC, etal. 2015. Abundant toxin-related genes in the genomesof beneficial symbionts from deep-sea hydrothermalventmussels. eLife 4 and analysis of the blue shark (Prionaceglauca) genomeSupervisorJudith Risse, Sandra SmitTypeData analysisRequirementsAdvanced Bioinformatics , GenomicsSkillsGenome analysis, genomics, PythonTimestampApril 2021 DescriptionCartilaginousfishes(Chondrich thyes)aredividedintotwosubclasses,elasmo branchs(Elasmobranchii,includingsharks,r aysandskates)andchimaeras(Holocephali), (2n=66 104).

5 Thissuggestshighgenomestability, , ,butalsomayprovidegenesofinterestwithbio technological (Prionaceglauca,genomesize~ ,2n=86)isacosmopolitanspecies,andmostfre quentlycaughtpelagicsharkspeciesinfishin gactivities,thereforeclassifiedas NearThreatened ,alongwiththeirhighecological,economic,a ndevolutionarysignificance, (~45xcoverage)andwill be assembled at the Leiden Centre for ,youwillannotatetheassembledgenomeusingP acBioIsoSeqdata(varioustissues) :1)researchingapproachestogenomeannotati onusingonlyPacBioHiFidata,2)structuralan notationofthebluesharkgenomeusingthesele ctedtools,3) powers ,ortheinvestigationoftissue-andhaplotype -specificgene-expressionpatternsmakingus eofthefulllength, ,youwillmakeuseofhigh-qualitydataproduce dwiththeextremely accurate PacBio HiFi (Keygene),KenKraaijeveld(Leiden),TiagoSi m es & Sara Novais (MARE, Portugal)ReferencesAlves et al.

6 Science of the Total Environment (2016) na et al. NAR Genomics and Bioinformatics (2021)3(1). & Ence. Nature Reviews Genetics (2012) 13(5). evolution of auxiliary metabolic genes in bacteriophagesSupervisorsAnne Kupczok, Marnix MedemaTypeData analysisRequirementsProgramming in python , Advanced BioinformaticsSkillsComparative genomics, databases, phylogenetics, python , statisticsTimestampSeptember 2020 DescriptionBacteriophages(short:phages) (AMGs), ( ).PhageswithAMGsareparticularlywellstudi edinthemarineenvironment,wherephagesarek nowntocontributesubstantiallytobacterial mortalityandwheretheyplayacentralroleinb iogeochemicalcycles( ).Phagesneedtopackagetheirgenomeintoapro teincapsidfortransmission, ,AMGsareoftenshorter, , ,AMGsmighthaveevolvedintodistinctphage-a daptedversions,whichalsohaveadeviatingfu nctionality( , ).Ontheotherhand,someAMGsmightbetransfer redbacktobacterialgenomes( , ).

7 Thesecasesprovideinterestingevolutionary scenarios,wheretheevolvabilityofbacteria isincreasedduetotheacquisitionofgene versions that evolved in ,publiclyavailablegenomesofphagesandbact eriawillbescannedformetabolicgenes( ).First, , back to bacterial M, Bonnain C, Malki K, Sawaya NA. puppet masters of the marine microbial :754 D, Sullivan MB, Johnson ZI, Tolonen AC, RohwerF, Chisholm SW. 2004. Transfer of photosynthesisgenes toand from Prochlorococcus :11013 B, Bowman Grahl S, Millard A, Corrigan RM,Clokie MRJ, Scanlan DJ. 2019. Cyanophage MazG isapyrophosphohydrolase but unable to hydrolyse magicspot Microbiology Reports11:448 M, Borton MA, McGivern BB, Zayed AA, RosaSLL, Solden LM, Liu P, Narrowe AB, Rodr guez-RamosJ,Bolduc B, et al. 2020. DRAM for distilling microbialmetabolism to automate the curation of J, Buchholz HH, Allen MJ, TempertonB.

8 2019. Host-hijacking and planktonic piracy: howphagescommand the microbial high Journal16 updates to a successful software forlarge-scale genomeminingSupervisorJorge Navarro, J r me Collemare,Marnix MedemaTypeData analysis, GenomeminingRequirementsAdvanced bioinformaticsSkillsProgramming ( python ),GenomicsTimestampApril 2021 DescriptionMicrobes and plants are able to synthesize secondarymetabolites (SMs) that allow them tothrive in their environment by offering means of communicationwith other organisms, defenseagainst competitors or environmental conditions, acquisitionof additional resources orfacilitating host colonization. The genes that makeup the biosynthetic pathways for thesemetabolites are co-regulated and, in microbes, oftenco-localized in the same genomic loci,generally referred to as biosynthetic gene clusters (BGCs).

9 The study of BGCs is of great relevance not only asa starting point to elucidate the bioactivitiesof the SMs whose production they encode, but alsoto predict the biosynthetic capacity ofnewly-sequenced genomes, to detect the presence ofknown harmful toxins and to characterizenovel pathways with potential to be of relevance tohumankind, such as those involved inproducing disease-treating drugs, pigments and cropprotection of regions containing putative BGCsis straightforward nowadays thanks tosoftware such as antiSMASH (1), which uses the principleof co-localization and knowledgeabout core biosynthetic genes those that synthesizethe scaffold of the metabolite. However,as genome sequencing is being scaled up, large groupsof BGCs need to be compared acrossgenomes, or within large sets of metagenome-assembledgenomes. Such large-scalecomparisons can help researchers discover BGCs thatare similar to characterized ones,avoiding time-consuming experiments (dereplication)or, on the contrary, to find interesting newversions of known BGCs.

10 Such analyses can also uncoverlarge groups of similar BGCs (genecluster families, GCFs) yet to be characterized, whichcan be prioritized for further tool that uses sequence similarity networks tocompare large datasets of BGCs is the biosynthetic gene similarity clustering and prospectingengine , or BiG-SCAPE (2), which usesconserved protein regions (domains) within predictedBGCs to carry out three different similaritycalculations that are combined into a final distancevalue for each pair of BGCs. By applying aclustering algorithm to each BGC subnetwork, BiG-SCAPE then groups BGCs into this project, you will work on bringing BiG-SCAPEto the next level by restructuring its codeand implementing new features to make it faster, lessresource-intensive, more accurate, andmore useful to researchers in the secondary (1) (2) content variation mediated by recombination inbacteriophageevolutionSupervisorsAnne Kupczok, Dick de RidderTypeData analysisRequirementsProgramming in python , Advanced BioinformaticsSkillsComparative genomics, databases, alignments, python , statisticsTimestampSeptember 2020 DescriptionBacteriophages(short:phages)a revirusesthatinfectbacteria, ( ).