DOMAIN: SYSTEMS BIOLOGY, EVOLUTION & POPULATION GENETICS
PIPELINE #086
Comparative Genomics, Orthology Inference & Gene Family Evolution
Technical Scope: Whole-proteome orthogroup clustering, species tree reconstruction, synteny conservation, and gene family expansion/contraction modeling.
Required Input Data & Formats
Complete Proteome FASTA Files (.faa) & Gene Coordinate BEDs (.bed) across 3 to 100+ Species.
Produced Output Files & Deliverables
Orthogroup Gene Matrix (.tsv), Ultrametric Species Phylogeny (.nwk), CAFE5 Expanded/Contracted Gene Families (.tsv), Macrosynteny Dot/Ribbon Plots (.pdf).
Core Software Toolchain & Frameworks
OrthoFinder 2, CAFE5, MCScanX / GENESPACE, MAFFT, TrimAl, IQ-TREE 2.
Key Literature & Methodological References
Emms & Kelly (2019) Genome Biology (OrthoFinder); Mendes et al. (2021) Bioinformatics (CAFE5); Lovell et al. (2022) eLife (GENESPACE).
Application
“Comparing genomes across species uncovers the evolutionary innovations behind unique traits—from venom biosynthesis and extremophile stress tolerance to crop disease resistance. Our Comparative Genomics pipeline integrates OrthoFinder 2 for accurate orthogroup inference, GENESPACE for chromosomal macrosynteny visualization, and birth-death stochastic modeling in CAFE5 to pinpoint rapidly expanding or contracting gene families. Gain deep evolutionary context to identify conserved functional cores and lineage-specific adaptations.”
DOMAIN: SYSTEMS BIOLOGY, EVOLUTION & POPULATION GENETICS
PIPELINE #087
Positive Selection & Molecular Adaptation (dN/dS) Rate Analysis
Technical Scope: Codon-aware alignment and site-, branch-, and branch-site likelihood ratio tests to detect episodic positive diversifying selection.
Required Input Data & Formats
Orthologous Coding DNA Sequences (CDS .fna) & Protein Sequences (.faa), Species/Gene Phylogenetic Tree (.nwk).
Produced Output Files & Deliverables
Codon-Aligned FASTA (.fna), Positive Selection Likelihood Ratio Test Table (.tsv), Positively Selected Codon Sites Mapped onto 3D Protein Structures (.pdf/.pml).
Core Software Toolchain & Frameworks
PAL2NAL, MACSE, HyPhy (aBSREL, MEME, BUSTED, FEL), PAML (codeml), EasyCodeML.
Key Literature & Methodological References
Kosakovsky Pond et al. (2020) Molecular Biology and Evolution (HyPhy); Yang (2007) Molecular Biology and Evolution (PAML); Suyama et al. (2006) NAR (PAL2NAL).
Application
“Signature patterns of elevated non-synonymous to synonymous substitution rates (dN/dS > 1) pinpoint exact amino acid residues driven by evolutionary arms races—such as viral receptor-binding interfaces, immune evasion proteins, and toxin diversification. Our Molecular Adaptation pipeline builds frameshift-aware codon alignments using MACSE and PAL2NAL, then executes branch-site and episodic selection models in HyPhy (MEME, aBSREL) and PAML codeml. We map positively selected residues directly onto 3D protein structures to guide functional mutagenesis and protein engineering.”
DOMAIN: SYSTEMS BIOLOGY, EVOLUTION & POPULATION GENETICS
PIPELINE #088
Population Genetics: Admixture, Demographic History & Selection Scans
Technical Scope: Population structure inference (PCA/ADMIXTURE), effective population size trajectory (PSMC/SMC++), and selective sweep detection (Fst, XP-EHH, iHS).
Required Input Data & Formats
Multi-Population Whole-Genome Phased VCF (.vcf.gz), Population Assignment Metadata (.tsv), Genetic Recombination Map (.map).
Produced Output Files & Deliverables
Ancestry Admixture Barplots & PCA (.pdf), Effective Population Size (Ne) Demographic Curves (.pdf), Selective Sweep Manhattan Plots & Candidate Genes (.tsv).
Core Software Toolchain & Frameworks
VCFtools, PLINK2, ADMIXTURE, SMC++ / Relate, selscan (iHS, XP-EHH), pixy, ANGSD.
Key Literature & Methodological References
Terhorst et al. (2017) Nature Genetics (SMC++); Szpiech & Hernandez (2014) MBE (selscan); Korunes & Samuk (2021) Molecular Ecology Resources (pixy).
Application
“Understanding population structure, historical bottlenecks, and recent selective sweeps is essential for conservation genomics, agricultural breeding, and human evolutionary genetics. Our Population Genetics pipeline computes unbiased nucleotide diversity (pi) and Fst using pixy, infers ancestry proportions via ADMIXTURE, reconstructs effective population size trajectories over millennia with SMC++, and scans for recent hard and soft selective sweeps using haplotype homozygosity statistics (iHS and XP-EHH) in selscan—uncovering the genetic loci driving local adaptation.”
DOMAIN: SYSTEMS BIOLOGY, EVOLUTION & POPULATION GENETICS
PIPELINE #089
Genome-Scale Metabolic Modeling (GEM) & Flux Balance Analysis (FBA)
Technical Scope: Reconstruction of organism- or tissue-specific metabolic networks, in silico gene knockout lethality screens, and media optimization.
Required Input Data & Formats
Annotated Genome (.faa) or Tissue Transcriptomics/Proteomics Matrix (.tsv), Reference Metabolic Models (BiGG, Human-GEM, AGORA2).
Produced Output Files & Deliverables
SBML Metabolic Model (.xml), Predicted Growth & Metabolite Secretion Fluxes (.tsv), Synthetic Lethal Gene Knockout Matrix (.csv), Escher Pathway Maps.
Core Software Toolchain & Frameworks
COBRApy, CarveMe, RAVEN Toolbox, Human-GEM / tINIT, Escher, memote.
Key Literature & Methodological References
Ebrahim et al. (2013) BMC Systems Biology (COBRApy); Machado et al. (2018) Nucleic Acids Research (CarveMe); Robinson et al. (2020) Science Signaling (Human-GEM).
Application
“Genome-Scale Metabolic Models translate a genome or transcriptomic profile into a mathematical stoichiometry matrix capable of simulating cellular metabolism in silico. Our GEM and Flux Balance Analysis pipeline reconstructs draft models via CarveMe or contextualizes Human-GEM using tissue RNA-Seq, validates model mass-balance with memote, and performs genome-wide single and double gene-knockout screens in COBRApy. Optimize bioproduction media in industrial biotechnology or predict cancer-specific metabolic synthetic lethalities before running costly wet-lab screens.”
DOMAIN: SYSTEMS BIOLOGY, EVOLUTION & POPULATION GENETICS
PIPELINE #090
Mendelian Randomization (MR) & Causal Genetic Target Validation
Technical Scope: Two-sample and multivariable Mendelian Randomization leveraging pQTL/eQTL genetic instruments to establish causal links between targets and disease.
Required Input Data & Formats
Exposure GWAS/pQTL/eQTL Summary Statistics (.tsv), Outcome Disease GWAS Summary Statistics (.tsv), LD Reference Panel (1000G).
Produced Output Files & Deliverables
Causal Effect Estimates (IVW, MR-Egger, Weighted Median, .tsv), Forest, Scatter & Leave-One-Out Sensitivity Plots (.pdf), Colocalization Posterior Probabilities (PP.H4).
Core Software Toolchain & Frameworks
TwoSampleMR, MendelianRandomization, coloc (Bayesian colocalization), MR-PRESSO, Steiger filtering.
Key Literature & Methodological References
Hemani et al. (2018) eLife (TwoSampleMR); Giambartolomei et al. (2014) PLOS Genetics (coloc); Verbanck et al. (2018) Nature Genetics (MR-PRESSO).
Application
“Drug targets supported by human genetic evidence are more than twice as likely to succeed in Phase II and III clinical trials. Our Mendelian Randomization and Colocalization pipeline leverages cis-pQTL and cis-eQTL genetic instruments as natural randomized trials to test whether modulating a candidate protein or metabolite causally alters disease risk. Combining TwoSampleMR, MR-PRESSO pleiotropy correction, and Bayesian colocalization in coloc, we separate true causal drug targets from spurious linkage disequilibrium associations—de-risking target selection at the earliest stage.”
DOMAIN: SYSTEMS BIOLOGY, EVOLUTION & POPULATION GENETICS
PIPELINE #091
Biomedical Literature Text-Mining & Knowledge Graph Extraction
Technical Scope: Automated Named Entity Recognition (NER), relation extraction, and ontology-grounded knowledge graph construction from PubMed and patent corpora.
Required Input Data & Formats
PubMed XML / Full-Text PDF Corpus or Search Query, Domain Ontologies (MONDO, ChEBI, Gene Ontology, HGNC).
Produced Output Files & Deliverables
Extracted Gene-Disease-Drug Relation Table (.tsv), Ontology-Grounded Knowledge Graph (.graphml / Neo4j CSVs), Evidence Sentence & PMID Citations.
Core Software Toolchain & Frameworks
PubTator 3.0 API, BERN2, scispaCy, BioBERT / PubMedBERT, NetworkX, Neo4j, LangChain.
Key Literature & Methodological References
Wei et al. (2024) Nucleic Acids Research (PubTator 3.0); Sung et al. (2022) Bioinformatics (BERN2); Neumann et al. (2019) BioNLP (scispaCy).
Application
“Over 36 million biomedical articles exist in PubMed, making manual literature curation for target discovery, drug repurposing, or safety signal detection impossible at scale. Our Biomedical Literature Text-Mining pipeline combines biomedical transformer models (PubMedBERT and BERN2) with ontology normalization across HGNC, MONDO, and ChEBI to extract structured gene-disease-drug relationships and causal mechanisms directly from scientific text. We assemble auditable, PMID-linked knowledge graphs that surface non-obvious therapeutic hypotheses and competitive intelligence.”
DOMAIN: SYSTEMS BIOLOGY, EVOLUTION & POPULATION GENETICS
PIPELINE #092
Survival Analysis & High-Dimensional Multi-Omic Prognostic Risk Modeling
Technical Scope: Time-to-event clinical outcome modeling integrating clinical covariates with high-dimensional genomic, transcriptomic, or proteomic predictors.
Required Input Data & Formats
Clinical Survival Data (Time, Event Status, Covariates in .csv), Normalized Multi-Omic Feature Matrix (.tsv).
Produced Output Files & Deliverables
Penalized Cox / Random Survival Forest Model (.rds/.pkl), Time-Dependent ROC & Calibration Curves (.pdf), Kaplan-Meier Risk Group Stratification Plots.
Core Software Toolchain & Frameworks
R (survival, survminer, glmnet, randomForestSRC, timeROC), Python (lifelines, scikit-survival).
Key Literature & Methodological References
Simon et al. (2011) Journal of Statistical Software (glmnet Cox); Ishwaran et al. (2008) Annals of Applied Statistics (Random Survival Forests).
Application
“Translating high-dimensional omics signatures into clinically actionable prognostic or predictive biomarkers requires survival models that handle censoring, tied events, and thousands of correlated features without overfitting. Our Survival and Prognostic Risk Modeling pipeline integrates ElasticNet-penalized Cox regression with non-linear Random Survival Forests under nested cross-validation, evaluating discrimination via time-dependent AUC and Brier calibration scores. Deliver validated patient risk scores that enrich clinical trial enrollment and demonstrate clear clinical utility.”
Back to Workflows