Paramvir S. Dehal

Contributions

Acinetobacter baylyi ADP1 Data Explorer analyzed a SQLite database of 15 tables, 461,522 rows and 135 MB for Acinetobacter baylyi ADP1 and 13 related genomes, and found that no single gene of the 5,852 in genome_features has data across all six modalities, although pairwise overlaps are substantial ₁. Triple Essentiality Concordance Analysis for *Acinetobacter baylyi* ADP1 compared flux balance analysis (FBA, constraint-based prediction of metabolic fluxes), TnSeq (transposon-insertion sequencing), knockout and growth data, and found that all 478 triple-covered genes were TnSeq-dispensable ₂. ADP1 Deletion Collection Phenotype Analysis analyzed 2,034 deletion mutants across 8 carbon sources and found mean growth ratios spanning a 3.3-fold range, from 0.41 on urea to 1.36 on quinate ₃. In that analysis, 97.9% of genes showed severe defects (growth ratio < 0.5) on urea, and only 1.6% were defective on quinate ₃. Aromatic Catabolism Support Network in ADP1 defined a 51-gene quinate support network around the β-ketoadipate pathway and, by co-fitness, assigned 44/51 genes (86%) to four functional subsystems plus 6 transcriptional regulators, with the regulators counted within the 44-gene total and 7 genes left unassigned because of low co-fitness with all subsystems ₄. Condition-Specific Respiratory Chain Wiring in ADP1 mapped 62 respiratory-chain genes across 8 subsystems and found that quinate requires only Complex I, whereas lactate specifically requires cytochrome bo3 ₅.

Conservation vs Fitness — Linking FB Genes to Pangenome Clusters mapped 44 of 48 Fitness Browser organisms to pangenome clades, producing 177,863 gene-to-cluster links, and found 27,693 putative essential genes (18.6%) across 33 organisms ₆. Fitness Effects vs Conservation — Quantitative Analysis and Gene Conservation, Fitness, and the Architecture of Bacterial Genomes reported a modest conservation gradient across 194,216 genes in 43 organisms, with essential genes 82% core and always-neutral genes 66% core _7,8. The synthesis also found that 24.4% of core genes, compared with 19.9% of accessory genes, show positive fitness when deleted ₈. Core Gene Paradox — Why Are Core Genes More Burdensome? found that this burden pattern is function-specific: it runs +12.9 percentage points for RNA Metabolism but -14.1 percentage points for Cell Wall ₉. The 5,526 Costly + Dispensable Genes characterized 5,526 costly and dispensable genes and found them dominated by mobile elements and poorly annotated DNA, though provenance was not measured directly ₁₀. COG Functional Category Analysis found that novel or singleton genes were enriched in COG (Clusters of Orthologous Groups) category L by +10.88% across 32 species ₁₁. The Pan-Bacterial Essential Genome identified 15 gene families essential in all 48 organisms and found that 859 of 17,222 ortholog families (5.0%) were universally essential ₁₂. Pan-bacterial Fitness Modules via Independent Component Analysis applied independent component analysis (ICA, a statistical decomposition into independent signals) across 32 organisms and found that it recovers co-regulated process modules, while ortholog transfer remains the stronger gene-level predictor ₁₃. Fitness Modules × Pangenome Conservation found that module genes were 86.0% core, compared with 81.5% for all genes (odds ratio, OR=1.46, p=1.6e-87) ₁₄. Co-fitness and Co-inheritance in Bacterial Pangenomes found that pairwise co-fitness only weakly predicts co-occurrence, and the across-organism Wilcoxon test was not significant (W=9, p=0.13) ₁₅.

Pan-Bacterial AMR Gene Landscape found that 30.3% of antimicrobial-resistance (AMR) genes were core, compared with 46.8% for the baseline (OR=0.49) ₁₆. Fitness Cost of Antimicrobial Resistance Genes found a pooled baseline AMR fitness cost of +0.086 [95% CI: +0.074, +0.098] across 25 organisms, of which only two remained significant after false discovery rate (FDR, the expected share of false positives among significant calls) correction ₁₇. Environmental Resistome at Pangenome Scale found that clinical species carry a median of 5 AMR clusters, compared with 2 for soil, aquatic and host-associated species, though the study is observational and subject to sampling, classification and annotation limitations ₁₈. Within-Species AMR Strain Variation found that 51.3% of 37,444 AMR gene-species records were rare, present in <=5% of strains within a species ₁₉. AMR Co-Fitness Support Networks found that AMR-containing ICA modules had a median of 46 genes, compared with 27 for non-AMR modules, and cautioned that enrichments may reflect shared dispensability ₂₀. Pan-Bacterial Metal Fitness Atlas found that metal-important genes were 87.4% core, compared with 76.9% for baseline genes (OR=2.08, p=4.3e-162) ₂₁. Gene-Resolution Metal Cross-Resistance Across Diverse Bacteria found that 98.1% (311/317) of gene-level metal fitness correlations were positive ₂₂. Metal-Specific vs General Stress Genes classified 4,177 of 7,609 metal-important records (54.9%) as metal-specific, meaning significant fitness defects under metal stress but a <5% sick rate across 5,945 non-metal experiments; the fraction was approximately 41% at a 2% threshold and approximately 67% at 10% ₂₃. That analysis could not process seven of 31 metal-tested organisms because their locusIds did not match the fitness-matrix index format, so the 24 included organisms accounted for 7,609 of 12,838 metal-important gene records (59.3%) ₂₃. Counter Ion Effects on Metal Fitness Measurements found that 39.8% of metal-important records were also NaCl-important, but found no evidence that chloride dose is the primary driver of the overlap ₂₄. BacDive Isolation Environment × Metal Tolerance Prediction found that heavy-metal-site isolates scored Cohen’s d = +1.00 above the baseline (p=0.006), but the n=10 group makes this estimate imprecise ₂₅. BacDive Phenotype Signatures of Metal Tolerance found that 7 of 10 BacDive phenotype features were significant after FDR correction ₂₆.

Lab Fitness Predicts Field Ecology at Oak Ridge found that 14 of 26 Fitness Browser genera were detected at Oak Ridge and that five genera were significantly associated with uranium, while aggregate tolerance was only suggestive ₂₇. Field vs Lab Gene Importance in *Desulfovibrio vulgaris* Hildenborough found that the deficits for antibiotic and heavy-metal genes in Desulfovibrio vulgaris Hildenborough were not significant after FDR correction, which leaves resistance accessory-enrichment as a single-organism hypothesis ₂₈. Contamination Gradient vs Functional Potential in ENIGMA Communities found, in confirmatory genus-level tests across 108 samples, no robust monotonic association between contamination and broad functional scores, while exploratory coverage-aware models detected positive defense associations that were sensitive to coverage, covariate specification, taxonomic resolution and multiple-testing correction ₂₉. Ecotype Correlation Analysis found a median environment partial correlation of 0.0025, compared with 0.0143 for phylogeny, across 172 species ₃₀. Ecotype Reanalysis — Environmental vs Human-Associated Species found that environmental species did not show stronger correlations than human-associated species (median 0.051 versus 0.084) ₃₁. Pangenome Openness Analysis detected no significant relationship between openness and environment effects (rho = -0.05, p = 0.54) ₃₂. AlphaEarth Embeddings, Geography & Environment Explorer characterized AlphaEarth embeddings for 83,287 genomes, 28.4% of 293,059 ₃₃. PaperBLAST Data Explorer — Literature Coverage Bias in Protein Sequence Space found that Homo sapiens accounts for 46.7% of PaperBLAST gene-paper records ₃₄. Web of Microbes Data Explorer characterized a 2018 Web of Microbes snapshot and flagged several internally inconsistent counts ₃₅. Metabolic Consistency of Pseudomonas FW300-N2E3 Across Four KBase Data Lakehouse Databases found that 17/21 testable metabolites (81%) were fully concordant across databases for Pseudomonas FW300-N2E3 ₃₆. Essential Metabolome: GapMind Pathway Analysis Across Essential-Gene Organisms found that 17 of 18 amino-acid pathways were present in all 7 organisms, in a pilot limited to 7 of 45 organisms ₃₇. Metabolic Capability vs Dependency classified 161 organism-pathway pairs and found 57 (35.4%) Active Dependency pairs, complete pathways containing fitness-important genes, and 66 (41.0%) Latent Capability pairs, complete pathways whose genes showed no significant fitness defects under standard conditions ₃₈.

Projects (38)

• acinetobacter_adp1_explorer
• adp1_deletion_phenotypes
• adp1_triple_essentiality
• amr_cofitness_networks
• amr_environmental_resistome
• amr_fitness_cost
• amr_pangenome_atlas
• amr_strain_variation
• aromatic_catabolism_network
• bacdive_metal_validation
• bacdive_phenotype_metal_tolerance
• cofitness_coinheritance
• cog_analysis
• conservation_fitness_synthesis
• conservation_vs_fitness
• core_gene_tradeoffs
• costly_dispensable_genes
• counter_ion_effects
• ecotype_analysis
• ecotype_env_reanalysis
• enigma_contamination_functional_potential
• env_embedding_explorer
• essential_genome
• essential_metabolome
• field_vs_lab_fitness
• fitness_effects_conservation
• fitness_modules
• fw300_metabolic_consistency
• lab_field_ecology
• metal_cross_resistance
• metal_fitness_atlas
• metal_specificity
• module_conservation
• pangenome_openness
• paperblast_explorer
• pathway_capability_dependency
• respiratory_chain_wiring
• webofmicrobes_explorer