EvidentialGene : Killifish Genes
 Name Last modified Size
![[DIR]](/icons/back.gif) Parent Directory 10-Jan-2014 14:45 -
![[DIR]](/icons/folder.gif) annotation/ 04-Jul-2015 21:46 -
current/ 13-May-2016 16:03 -
![[TXT]](/icons/text.gif) gene_summary.txt 09-Jan-2014 13:38 15k
inotherfish/ 13-May-2016 00:53 -
kf2genesearch.html 09-Jan-2014 17:09 4k
kfish2rae5/ 13-May-2016 16:03 -
kfish2submit/ 14-Sep-2015 13:43 -
mRNA_assembly/ 04-Dec-2013 13:18 -
modelled_on_genome/ 21-Oct-2013 01:47 -
old_versions/ 04-Jul-2015 21:53 -
# kfish2rae5g_sum.txt
2013-Nov-12++
Killifish, Fundulus heteroclitus genome project
http://arthropods.eugenes.org/EvidentialGene/killifish/project/
Gene assemblies, v2 an5g, 2013 December
at killifish/project/Genes/kfish2rae5/
Summary in killifish/project/Genes/gene_summary.txt
kfish2rae5g version files are separated in main (primary transcript) and alt (alternate) parts
Genome assembly v2.1, killifish201303asm
at killifish/project/Genome/
Gene annotation: Search at killifish/project/ bottom form or
killifish/project/version2/genes/kf2genesearch.html
Gene families,Fish orthology: Search at killifish/project/ (FISH11G)
Genome maps: at Genome v2b (2013 Mar) at killifish/project/
BLAST search genes, genome: killifish/project/BLAST/
Gene evidence:
killifish/project/Introns/
347K valid introns on genome asm from RNA,EST reads
killifish/project/Proteins/
proteins of 9 fish species + human used for orthology evidence (n=284956 main isoform set)
killifish/project/RNAs/
kf2evg367mixx11/ : mRNA assembly, 183K used of 723K, from 776M short reads, 65 Bil bases, see below
est_cgbAssembly100.fa : 57741 assemblies of 3.5M 454-reads, from 3 projects 2009-2012
killifish/project/Repeats/
Repeatmasker transposon and repeat finding, and masked genome assmbly (12% masked)
---------------------------------------------------
TABLE G1. Gene set numbers, version kfish2rae5d, 20 Nov 2013
---------------------------------------------------
35047 gene loci, all supported by mRNA transcripts and/or protein homology evidence
27597 from mRNA-assembly, 4929 genome-modelled, 2514 Kfish version 1 (mixed sources)
21083 are orthologs to other species, 3701 are inparalogs of orthologs, and 10263 are species-unique.
25753 properly-mapped to genome (>=80% coverage),
2292 un-mapped genes, 2328 split-mapped, 4694 partial-mapped <80% coverage
1851 single-exon loci (including partials, excluding TE)
79103 alternate transcripts at 20449 loci, ave. 5 transcripts per locus, 40 maximum.
27069 have complete proteins, 7966 have partial proteins
2504 have <90% evidence coverage, 1690 have <60%, by orthology and/or RNA evidence
2460 Transposon-associated expressed transcripts
---------------------------------------------------
Gene set numbers, version kfish2rae5g, 15 .. 27 Dec 2013
34928 gene loci
- a few changes vs an5f, mostly seq corrections (antisense aa, seleno), same kfish2rae5f.gff?
- recovered 28 kfish1 loci not in kf2, strong orthology (3 missed, 19 improved fish ortho grps)
- omcl updated for new aa seqs
- 3500 alternate transcripts removed due to conflicting locus info are
otherwise valid and should be retained, but confict noted to be resolved
(two-loci or wrong-locus)
* Note that some conflicts in alt-paralog classing remain (both ways).
Gene set numbers, version kfish2rae5f, 04 Dec 2013
34903 gene loci
- minor number changes since rae5d, 100 dup-loci removed, 300 loci updated
- mostly resolves discrepancies between sequences and gff locations, matching object ids,
---------------------------------------------------
TABLE G2a. Fish species average orthology gene groups
---Common Groups--- ----All Groups-----
Species cBits aaSize orMiss tBits orGroup Tiny
--------- -------------------- --------------------
killifish 803 50 18 585 17272 1.1%
maylandia 824 45 76 596 16469 1.1%
tilapia 822 6 223 568 14905 1.9%
platyfish 783 -12 118 549 15305 4.7%
zebrafish 711 -9 366 478 15190 4.8%
sticklebk 763 -42 342 509 14343 7.8%
catfish 725 21 729 470 14276 3.4%
medaka 743 -45 654 478 13541 9.7%
tetraodon 732 -50 658 473 13423 7.9%
spotgar -- -110 2588 329 10882 20.2%
human -- 27 -- 395 12606 1.7%
----------------------------------------------------
source: kfish2/prot/fish11c/fish11gor3, Dec11 .. fish11gor3 update
cBits = bitscore average for 8656 common fish gene groups
tBits = bitscore average for all ortholog groups
aaSize = average protein size difference from group median
orMiss = missing ortholog groups that are common to other 8 of 9 fish (-gar)
orGroup = number of ortholog gene groups in species
Tiny = percent species gene size outliers below 2sd of group median size
----------------------------------------------------
Killifish 2013/2012 improvements (relative to tilapia)
Geneset cBits dSize orMiss tBits orGroup Tiny
--------- -------------------- -------------------
kfish.2v1 +12 +48 -100 +40 +1690 -3.6% : all improvements
---------------------------------------------------
TABLE G2b. Fish gene orthology categories (using OrthoMCL)
----------- GENES ----------- ------ GROUPS ----- Ortho to Kfish
nGene Orlog Inpara Uniq1 UDup OrGrp OrMis1 UniGrp Shared Best
----------------------------- ------------------- -------------
killifish 34931 21133 3672 7694 2432 17272 10 682 --- ---
maylandia 23194 21021 879 1171 122 16469 46 52 15468* 5159
tilapia 21437 19461 975 810 189 14904 135 78 14019 954
platyfish 20366 19483 214 641 29 15307 54 14 14609* 13352
zebrafish 26190 18465 4089 2202 1439 15187 188 226 13286* 1363
stickleback 20787 17954 1254 1396 181 14344 180 44 13317* 636
catfish 43671 17279 2964 15470 7938 14246 407 1561 12799* 1208
medaka 19686 16881 912 1535 361 13542 366 106 12670* 1071
tetraodon 19602 16814 904 1796 176 13425 360 67 12475* 696
spotgar 15734 11841 2655 1009 230 10880 1514 56 9492 116
human 39357 12699 7758 12497 6402 12608 420 2221 11265* 1182
-------------------------------------------------------------------------------
source: kfish2/prot/fish11c/fish11gor3, Dec11 .. gor3 upd Dec27
nGene = count of input genes, excludes alternate isoforms/locus.
Orlog = Orthologous genes (one-to-one matches among species)
Inpara = Inparalogs (recent ortholog duplicates) of orthologous genes
Uniq1,UDup = single-copy and duplicated species-unique genes
OrGrp,UniqGrp = orthologous and species-unique groups
OrMis1 = groups missing in species that all other species have
Ortho to Kfish, Shared= count of ortho groups shared with killifish,
Best = count of groups with closest homolog,
Shared* = maximum shared of 10 choices, tilapia shares more with maylandia,
and spotgar with zebrafish.
------------------------------------------------------------------
TABLE G2c. Fish Taxonomy with Human gene alignment stats
Human genes Fish Taxonomy
-------------------------------------------------------------------------------
Nhuman Ident% Align% Neopterygii
+ Teleostei
+ + + Euteleostei
+ + + + + + Pseudocrenilabrinae
14822 66 71 :+ + + + + + + + Maylandia zebra # african cichlid Zebra Mbuna, NCBI : KF2
14181 65 70 :+ + + + + + + + Oreochromis niloticus # Tilapia, Ensembl : KF1,2
+ + + + + Smegmamorpha
14893 64 70 :+ + + + + + + Gasterosteus aculeatus # Stickleback, Ensembl? : KF1,2
+ + + + + + Atherinomorpha
14478 64 67 :+ + + + + + + + Oryzias latipes # Medaka, Ensembl? : KF1,2
+ + + + + + + Cyprinodontiformes
15033 64 71 :+ + + + + + + + + + Xiphophorus maculatus # Platyfish, Consortium : KF2
17072 65 76 :+ + + + + + + + + + Fundulus heteroclitus # Killifish, Evigene : KF1,2
+ + + Otocephala
16127 65 70 :+ + + + + + Ictalurus punctatus # Catfish, Evigene : KF2
16871 64 73 :+ + + + + + Danio rerio # Zebrafish, Consortium/Ensembl : KF1,2
+ Semionotiformes
13081 70 54 :+ + Lepisosteus oculatus # Spotted gar, Draft Ensembl : KF2
-------------------------------------------------------------------------------
for N =26859 human genes, nc=16555 common to 7+ fish for align score
TABLE G3a. Gene Expression (RNA-seq) summary
---------------------------------------------------
34641 (99%) loci have FPKM >= 0.02 (~20+ reads/1000 bases)
29139 >= 1 FPKM, 14882 >= 10 FPKM
7 median, 58 mean, 20000 max FPKM
Approx. differential expression
nodiff :21881, 63.5%
adultenv: 5707, 16.5%
embryo : 4666, 13.5%
grandis : 2185, 6.3%
---------------------------------------------------------------------
RNA-seq read groups are whoi (embryo, 106M 2x100b read-pairs, 56% mapped),
mdibl (adult tissue/environ stress, 106M 2x100b read-pairs, 59% mapped),
and grandis (Fund. grandis sibling species, 176M 2x60b read-pairs, 67% mapped)
total of 388 million read pairs, 65 billion bases.
bowtie-mapped to CDS sequences [ kfish2rae5g.mainalt.rnax3.tab ]
FPKM : fragments mapped to transcript, per sequence kilobase, per million mapped reads.
TABLE G3b. Orthology, SNP by DE group effects
unique inpar snp
adultenv 35%+ 31%+ 40%
embryo 11%- 7%- 34%-
grandis 38%+ 23% 55%+
nodiff 31% 12% 41%
---------------------------------------------------------------------
unique = species-specific genes versus ortholog genes
inpar = Inparalogs (new genes) versus orthologs
snp = presense/absense of read SNPs mapped to transcript
-----------------------------------------------------------------------
TABLE G3c. Gene functions (GO terms) with most divergent
differential expressed responses
-----------------------------------------------------------------------
GOntol_ID %adlt adult embr grnd none GO_Term
-----------------------------------------------------------------------
GO:0046906 90.0 63 21 38 70 F:tetrapyrrole binding
GO:0030246 82.9 58 13 18 70 F:carbohydrate binding
GO:0008289 55.3 26 13 10 47 F:lipid binding
GO:0016491 53.0 221 48 98 417 F:oxidoreductase activity
GO:0033218 51.2 22 3 8 43 F:amide binding
GO:0060090 50.6 42 33 2 83 F:binding, bridging
GO:0048037 38.4 78 22 44 203 F:cofactor binding
GO:0043021 38.1 16 1 1 42 F:ribonucleoprotein complex binding
GO:0043178 36.0 18 21 14 50 F:alcohol binding
..
GO:0022610 54.6 311 463 49 570 P:biological adhesion
GO:0051704 53.7 341 81 54 635 P:multi-organism process
GO:0050896 41.5 1208 713 302 2911 P:response to stimulus
GO:0043170 37.4 79 57 20 211 P:macromolecule metabolic process
GO:0044255 36.4 223 94 103 612 P:cellular lipid metabolic process
GO:0048469 36.2 21 24 3 58 P:cell maturation
-----------------------------------------------------------------------
TABLE G4. Evigene mRNA assembly set,
project/RNAs/kf2evg367mixx11/kfish2evg367mixx11pub
mRNA inputset: 723555 kfish2evg367mixx from subset assemblies:
356953 Funhe2E6b, 175596 Funhe2Eq7, 136132 Fungr1EG3, 54874 Funhe2Emap3
mRNA classification:
Class cull drop okay Class notes
----------------------------------------------------------------------
althi 0 35078 51301 # high identity exon alternate
althi1 0 122038 53038 # higher idenity alternate
altmap 0 8741 10828 # alternate from genome mapping
altmid 0 10694 10413 # mid identity alternate (may be paralog)
main 0 1713 45048 # main transcript, longest with alts
mainsingle 0 40406 9629 # main transcript, no alternate
frag1exon 16482 0 0 # fragment single exon, no homology
fragalthi 0 65834 0 # fragment high ident alternate
fragaltmid 0 21182 3025 # fragment mid ident alternate
fragnearg 31015 0 0 # alt-exon near gene, but unattached
fragtrivia 85597 0 0 # trivially short, no homology
fragnopath 6342 0 0 # short, no genome map, no homology
----------------------------------------------------------------------
total 139436 305686 183282
okay= keep, drop= uninformative excess, cull= uninteresting excess
Notes:
-------------
Killifish, Maylandia and Tilapia form a good gene methods/results
comparison, as top-scored gene sets, recently built by 3 groups with
"good" gene construction pipelines. Some artifacts of methods may be
found. Tilapia has Ensembl:genewise+exonerate models from mix of
rna-seq + uniprot prots. Mayzebr is NCBI Gnomon annotate, also mix of
RNA-seq and related proteins, Kfish2 is Evigene annotate, mRNA-gene
strong but also using related species proteins. Other two do use
RNA-seq assembly, but not as extensively or carefully, and rely on
mapping to genome assembly. Ensembl-genewise-protein mapping has
potential to add artifacts of homolog models. NCBI Gnomon now uses
RNA-seq more carefully than in past, and better than Ensemble I think.
Killifish and Platyfish form another useful comparison, platyfish being closest
relative, and also a recent genome product built with current data and software.
Differences that can be highlighted:
1. "The quality of a gene set is dependent on the quality of the genome assembly"
(from Ensembl platyfish gene build document). This also can be derived from methods
of platyfish genome paper, e.g. the methods included discarding mRNA assemblies
that did not map well to genome assembly).
In contrast, killifish genes v2 are not dependent on quality of genome assembly,
merging both mRNA-assembly and genome-mapped methods to pick best set from both.
2. The human gene orthology stats indicate killifish surpases platyfish in
completeness of genes.
Killifish, Maylandia and Catfish form a third special comparison to other
fish genes. You will find in the Orthology search that these three share
more gene families that are missed in the other fish, than any other 3-fish
comparison, by about 100 families. This is I think an effect of (a) mRNA
assembly independent of genome genes used for Killifish and Catfish, and (b)
for Maylandia, the NCBI has improved its mRNA evidence use enough to be
roughly equivalent to discovering genes that may be poorly modelled on genome
assembly.
Find families shared by just these 3 fish,
http://arthropods.eugenes.org/lucegene_arthropod/search?q=fish11xml-all:geneid+AND+Killifish:[1+TO+999]+AND+Catfish:[1+TO+999]+AND+Maylandia:[1+TO+999]+AND+Medaka:0+AND+Stickleback:0+AND+Tetraodon:0+AND+Tilapia:0+AND+Zebrafish:0+AND+Platyfish:0
One of these is http://arthropods.eugenes.org/genepage/fish11xml/FISH11G_G18567
FISH11G_G18567 : new D-tyrosyl-tRNA(Tyr) deacylase, one of 3 same named families,
in killifish, catfish and maylandia only, maylandia: XP_004554729.1/LOC101478506, 1035 aa
FISH11G_G1773 : D-tyrosyl-tRNA deacylase member 2, all but killifish, various number of genes (Tetraodon has 10)
maylandia: XP_004554728.1/LOC101478225, 168 aa
G1773 and G18567 are related in that G1773 shorter protein aligns to longer G18567,
both have same CDD:202294 domain. In maylandia, these are tandem genes.
In killifish, missing shorter one would be where genome gap exists (mRNA assembly may or
may not have partial version)
FISH11G_G5675 : D-tyrosyl-tRNA deacylase member 1, one gene in all 11 species
|