### Aug 25 2011 ###
Reading list
Laessig and Mustonen, new concept of selective flux
Pritchard 2001 and Reich and Lender’s opposing population genetic modeling for common disease common/rare variants
### Apr 30 2011 ###
Ionita-Laza, I., Buxbaum, J. D., Laird, N. M., Lange, C., Feb. 2011. A new testing strategy to identify rare variants with either risk or protective effect on disease. PLoS Genet 7 (2), e1001289+. URL http://dx.doi.org/10.1371/journal.pgen.1001289
[Notes]
The basic idea of the paper: “The strategy is based on assessing whether rare variants in a genetic region collectively occur at significantly higher frequencies in cases compared with controls (or vice versa).”
It is claimed that previous methods based on weighted sum of rare variants are sensitive to the presence of both protective and risk variants. In this method, one could separately test the two cases, or assume that both exist and use a two-sided test, which cause some loss of power.
### Oct 15 2010 ###
Liu, D. J. & Leal, S. M. A novel adaptive method for the analysis of next-generation sequencing data to detect complex trait associations with rare variants due to gene main effects and interactions. PLoS Genet 6, e1001156+ (2010). URL http://dx.doi.org/10.1371/journal.pgen.1001156.
[Methods]
Two types of main effects phenotypic model are considered: 1.) constant genetic effects for each causal variant and 2.) genetic effects inversely correlated with minor allele frequencies (MAF) of causal genetic variants.
Also evaluated two gene interaction models: 1). within gene interaction between promoter and multiple nonsynonymous; 2) between genes, where rare variants in different genes which are in a shared pathway, where additional variants don’t increase the risk on background of the previous ones
### Aug 18 2010 ###
Maximum parsimony vs likelihood based method (Kern & Begun 2005 MBE)
Parsimony and maximum-likelihood analyses were performed on a site-by-site basis to infer ancestral states for the common ancestor of D. melanogaster and D. simulans. Inferred ancestral states were used to identify lineage- specific fixations and to infer ancestral states for poly- morphisms. For likelihood reconstruction, we used PAML (Yang 1997) under a number of substitution models. For these reconstructions, we assumed that within-species genealogies are star shaped. This could be a problem if one were indirectly inferring the number of polymorphisms within species because the assumption of a star-shaped genealogy would inflate the number of mutations inferred on terminal branches. However, because we are interested only in the internal node of the genealogy, which represents the most recent common ancestor of D. melanogaster and D. simulans, the star-phylogeny assumption is not expected to have a large effect on our inferences. Indeed, any error caused by this assumption appears to have had little effect on our conclusions, as we obtained qualitatively similar results independent of the mode of ancestral reconstruction employed. Furthermore, only minor differences were ob- served for different likelihood models (data not shown), most plausibly as a function of the short divergence times. Results reported here are from maximum-likelihood re- constructions under the general time reversible (GTR) model (Tavare ́ 1986), where both the transition/transversion rate and shape parameter for rate heterogeneity were esti- mated from the data. For parsimony reconstructions, only sites for which an unambiguous ancestral state could be inferred were considered. Although parsimony may lead to inaccurate ancestral inference over large evolutionary dis- tances, it should provide a reliable criterion over the short divergence times between taxa considered here (Yang, Kumar, and Nei 1995). All parsimony analyses were done using software written by A.D.K.
GC-biased gene conversion contributing to SFS and MK table (Ometto et al 2006, Biol. Lett.)
- Polarizing a total of 1920 and 1564 SNPs in INs and IGs, respectively, shows that GC nucleo- tides exhibit a stronger tendency to mutate to AT than vice versa (p!0.0001; table 2). The analysis of the frequency spectra revealed that AT/GC poly- morphisms segregate at a significantly higher average frequency (0.291G0.009;G1 s.e.) than GC/AT ones (0.256G0.006; p!0.001). This is also found when INs and IGs are considered separately (figure 1 of electronic supplementary material).
- in Intergenic regions, recombination rate shows a significant positive correlation with the frequency of AT->GC polymorphism (r=0.099, p=0.042) and a negative correlation with GC content (r=-0.251, p=0.015)
- The first observation suggests a recombination-associated fixation bias of GC polymorphism. If so, the observed negative correlation between GC content and recombination rate indicates that this bias is counteracted by some other force
### Aug 17 2010 ###
EVE-GAL4, for stripe-specific-chip
McKay, L. M., Carpenter, B. & Roberts, S. G. E. Evolutionary conserved mechanism of transcriptional repression by even-skipped. Nucl. Acids Res. 27, 3064-3070 (1999). URL http://dx.doi.org/10.1093/nar/27.15.3064.
Our data support the hypothesis that Evecontainsan active repression domain that functions specifically to preventpreinitiation complex formation.
### May 20 2010 ###
[Major Point] Higher order interactions
using SELEX-SAGE coupled with a HMM analysis, they were able to survey the impact of higher order interactions. Their conclusion for CTF (TF) is that 1) adjacent base pairs are the most important. 2) higher order interactions are significant but contribute no more than 1 bit of information in a 12-bit motif. So the expected impact is relatively small.
[Note]
What I found as the most dramatic effect is the plateau in multioff sequences, i.e. after ddG up to 4kcal/mol from the optimum, dG no longer decreases, suggesting the protein switch to the non-specific binding mode. This is not the same as higher order interactions.
Roulet, E. et al. High-throughput selex sage method for quantitative modeling of transcription-factor binding sites. Nature biotechnology 20, 831-835 (2002). URL http://dx.doi.org/10.1038/nbt718
