Review papers with raw data transparency

Paper: Kim J., "Macro-Evolution of the hairy Enhancer in Drosophila species" (Journal of Experimental Zoology, 2001) — DOI: 10.1002/jez.1067 Kim 2001 — hairy enhancer

Goal: test whether stabilizing selection preserves spacing (length) between conserved binding-blocks of the Stripe 1–5 hairy enhancer across species.

Key empirical observations (paper-reported)

• Enhancer architecture: alternating short conserved binding-blocks and long poorly alignable variable regions; conserved blocks often include known/putative gap-factor motifs (hb, Kr, bcd, cad, antp, prd/ftz/eve/en) Kim 2001 — enhancer structure
• High indel density in variable regions, multiple lineage-specific disruptions of conserved blocks (e.g., Kr and hb motif changes; D. pseudoobscura lineage especially divergent) Kim 2001 — lineage changes
• Phylogenetic trees from conserved blocks, variable blocks, and full data match accepted relationships for close taxa but fail at deeper scale; molecular-clock rejected for combined and conserved sets (log-likelihood ratio tests reported) Kim 2001 — phylogeny & clock tests
• Statistical test for stabilizing selection on inter-block length: hierarchical F-test on covariances (adjacent segments -> larger groups) — result: cannot reject null hypothesis of independent variation; only weak/patchy negative covariance signals (segments 1–4 and 12–16) and positive covariances near 3' end (segments with similar TF motifs) Kim 2001 — stabilizing selection test

Critical appraisal — strengths and limitations

• Strengths: careful primary sequencing across multiple species; explicit partitioning into conserved vs variable regions; phylogenetic analyses and a statistical framework for testing covariance/stabilizing selection are implemented and reported with numbers (LRTs, F-statistics) Kim 2001 — methods/results
• Limitations & potential blindspots:
  • Small taxon sample for a macro-evolutionary claim (7 species) — reduces power to detect stabilizing selection acting across broader clades (Kim acknowledges macro vs micro distinction) Kim 2001 — sampling note
  • Alignment ambiguity in variable regions: CLUSTAL W + manual adjustment used, but indel-rich regions are "barely alignable" — uncertain homology of length changes can bias covariance estimates (Kim documents alignment difficulties) Kim 2001 — alignment ambiguity
  • Statistical test assumptions: the F-test used assumes independent species samples and normally distributed length variation; Kim acknowledges phylogenetic nonindependence (paired close taxa) and that the test is "liberal" for presence of negative covariance — degrees-of-freedom adjustments suggested but not fully resolving nonindependence; modern phylogenetic comparative methods (PGLS, phylogenetic eigenvectors) would better control shared history (not available in 2001 analysis) Kim 2001 — statistical caveat
  • Functional link between length variation and phenotype remains inferential: while spacing affects cooperative TF binding in model systems (cited in paper), direct experimental tests (transgenic swaps, reporter assays) for Stripe1–5 enhancer function across species are needed to validate selection inference (Kim cites Ludwig et al. 1998/2000 and functional literature but does not perform transgenics here) Kim 2001 — functional inference limit

Conclusions, confidence, and falsifiability

Kim's main empirical claim — absence of strong macro-evolutionary stabilizing selection on inter-block lengths in the Stripe1–5 hairy enhancer — is supported by the data and the F-test framework but remains moderate-confidence because (1) taxon sampling is small for macro-evolutionary claims, (2) alignment/ homology ambiguity in indel-rich variable regions weakens precision, and (3) the statistical test does not fully account for phylogenetic nonindependence using modern comparative methods. Evidence that would overturn Kim's conclusion: (A) broader sampling across Drosophila clades showing consistent negative covariance across many independent lineages; (B) functional reporter assays (transgenic cross-species swaps) demonstrating that particular segment-length changes measurably reduce fitness or stripe accuracy; or (C) phylogenetically-corrected comparative analyses revealing significant negative covariances after controlling for shared ancestry Kim 2001 — conclusion caveats