Hi Brooke,
thanks for looking into the font situation.
couple more things:
(1) be good if it had publication-quality output options. for example, the main
browser graphic has pdf/ps output options.
i have not looked under the hood of the tree utility software, but the final
web page product appears to be a raster graphics file in .png format [that the
utility intro page calls a .gif and the page url calls phyloGif]. somewhere
along the way I think it generates a vector file (adobe illustrator-like) for
both the tree part and spliny text. it rasterizes that to some internal size
and then expands it to the user-specified width and height (pixellating the
text if a large size is specified for a small tree). mac firefox also rescales
it to fit in the window unless wrap in html box is checked. so we just need to
capture the vector file along the way.
2. It would be desirable to retain smooth scalability and text-editing. One
problem it has now is in labeling internal divergence nodes (eg primates). The
labels get too crowded and text piles up on lines and the end leaf text. One
important special case is labelling of outer ladder nodes. Here the higher
order taxa labels could go on the outside. There is one label for every
parenthesis in the example tree, namely (((((((((((((((((((((((((((((homSap.
These represent the reconstructable ancestral sequences on the direct
evolutionary trajectory to human (a big goal of tree-drawers). the attached pdf
shows a few outside labels i put in by hand. these are non-interfering with
tree and leaf text, except maybe altering total width.
Here is a big tree to illustrate issues. Big trees (say 250 species) are the
future because of the many sequencing projects. It's very desirable to present
these trees compactly.
(((((((((((((((((((((((((((((homSap,panTro),gorGor),ponAbe),nomLeu),(macMul,papHam)),(calJac,saiBol)),tarSyr),(otogar,micMur)),tupBel),((((((musMus,ratNor),(criGri,(perMan,perPol))),dipOrd),cavPor),speTri),(oryCun,ochPri))),(((vicPac,(susScr,(turTru,(oviAri,bosTau)))),(((felCat,(canFam,ailMel)),equCab),(myoLuc,pteVam))),(eriEur,sorAra))),(((loxAfr,proCap),echTel),(dasNov,choHof))),((macEug,monDom),sarHar)),ornAna),((((((galGal,melGal),lagLag),anaPla),taeGut),croPor),anoCar)),(ambMex,((xenTro,xenLae),ranCat))),latCha),(((salSal,oncMyk),(esoLuc,(gadMor,(tetNig,((hipHip,solSen),(oreNil,(dicLab,(perFla,spaAur)))))))),(ictPun,((cteIde,cypCar),danRer)))),(leuEri,squAca)),hagfish),(((cioInt,cioSav),(halRor,botSch)),oikDio)),braFlo),((((strPur,parLiv),lytVar),patPec),(sacKow,(((a,b),c),d)))),((((((a,b),c),d),(((a,b),c),d)),(((a,b),c),d)),(((a,b),c),d))),(((((a,b),c),d),((((a,b),c),d),(((a,b),c),d))),(((a,b),c),d))),(((a,b),c),d)),(((a,b),c),d)),(((a,b),c),d)),monBre);
3. striping. here some sort of alternating color striping of the ancestral
nodes would be helpful for visualizing the subclades of the main ancestral
nodes described above. These are of variable height depending on how many
species are in the subclade (eg two for old world monkey, 10 for rodents +
lagomorphs). i don't have photoshop handy but took a screenshot to illustrate
concept. they could just be bands across the entire width of the output, tree
and text.
Thanks!
Tom
On Dec 2, 2011, at 3:32 PM, Brooke Rhead wrote: Hi Tom,
Thank you for bringing this issue with the fonts in the tree drawing utility to
our attention. We have added the request our tracking system. I don't have
any time frame yet on when this might be implemented. Brooke Rhead
UCSC Genome Bioinformatics Group
(((((((((
(human_hg18:0.00669,chimp_panTro1:0.00757):0.0243,
macaque_rheMac2:0.0592):0.0240,
((rat_rn4:0.0817,mouse_mm8:0.0770):0.229,
rabbit_oryCun1:0.207):0.107):0.0230,
(cow_bosTau2:0.159,dog_canFam2:0.148):0.0394):0.0285,
armadillo_dasNov1:0.150):0.0160,
(elephant_loxAfr1:0.105,tenrec_echTel1:0.260):0.0404):0.218,
monodelphis_monDom4:0.371):0.189,
chicken_galGal2:0.455):0.123,
xenopus_xenTro1:0.782):0.156,
((tetraodon_tetNig1:0.199,fugu_fr1:0.239):0.493,
zebrafish_danRer3:0.783):0.156);
_______________________________________________
Genome maillist - [email protected]
https://lists.soe.ucsc.edu/mailman/listinfo/genome
