What's the longest functional transcript known? I'm wondering about RNA length post splicing, so not including introns.

  • $\begingroup$ You are looking for the longest, processed mRNA transcript. In which species? Human? $\endgroup$
    – Chris
    Sep 15 '14 at 21:48
  • $\begingroup$ I was wondering if anyone knew of any beyond the human set. We might have found one today in the Databases.... $\endgroup$
    – shigeta
    Sep 15 '14 at 22:30

Top 10 long processed transcripts in humans (with multiple isoforms), from gencode 19 annotations:

Transcript    Length(bases)
TTN-018        108861     <-- Titin
TTN-019        103988      
TTN-002        101206     
KCNQ1OT1-001    91666
TTN-201         82413
TTN-202         82212
TTN-003         81838
MUC16-001       43732
TSIX-001        37026
MCC-009         29616

Ignoring isoforms (only longest isoforms shown)

Transcript    Length(bases)
TTN-018         108861
KCNQ1OT1-001     91666
MUC16-001        43732
TSIX-001         37026
MCC-009          29616
TRAPPC9-015      29514
SYNE1-001        27602
GRIN2B-001       27204
OBSCN-011        26811
NEB-204          26020

Titin clearly is the longest transcript in humans

However this is the list of longest genes:

Gene           Length(Kb)
CNTNAP2        2304.64
LSAMP          2186.93
DLG2           2169.35
DMD            2092.29
PTPRD          2084.57
MACROD2        2057.83
CSMD1          2056.87
EYS            1987.24
LRP1B          1900.28
PCDH15         1806.76
CTNNA3         1783.65
ROBO2          1740.82
RBFOX1         1691.87
NRXN3          1619.64
DAB1           1548.83
RP11-420N3.2   1536.21
PDE4D          1513.42
FHIT           1502.09
AGBL4          1491.06
CCSER1         1474.33

Top 5 in Zebrafish (Zv9.75); longest isoforms:

ttna-203             93727   <-- Titin
ttnb-202             82632
si:dkey-16p6.1-001   67263
syne2b-201           31867
si:dkey-30j22.1-001  29269

Top 5 in Drosophila (FlyBase r6.02); longest isoforms:

dp-RQ           71300   <-- Dumpy
sls-RP          56448   <-- Titin
Muc14A-RA       48719
Msp300-RG       43105
Ank2-RU         42107

Top 5 in C.elegans (WormBase WS220); longest isoforms:

W06H8.8g        55623   <-- Titin
K07E12.1a.2     39257   <-- dig-1
ZK973.6         25608
C09D1.1b        24198
C41A3.1         23457

Top 5 in Arabidopsis (TAIR 10.23):

AT1G67120.1     16272   <-- Midasin homolog
AT3G02260.1     15451   <-- Calossin-like protein
AT5G28263.1     15194
AT1G43060.1     14622
AT5G30269.1     14590

Top 5 in yeast (SGD):

YLR106C         14733   <-- Midasin
Q0045           12884   <-- Subunit I of cytochrome c oxidase
YKR054C         12279
YHR099W         11235
YDR457W          9807
  • 1
    $\begingroup$ It seems that Titin is on the top of the list in many species. Interesting. $\endgroup$
    – Chris
    Sep 16 '14 at 8:02

I think a good candidate is the human titin gene. The gene itself has 363 exons, depending on the isoform it has between 27.000 and 34.000 residues. This makes up a processed mRNA length of up to 100kb for the full length isoform. See either the Wikipedia article or the one linked below for more details:

If you are looking for the longest primary transcript, then the human dystrophin gene should be your favorite. It has a length of about 2.4 Megabases for the primary transcript, the processed transcript has only a length of about 14kb. It takes about 16 hours to transcribe and co-transcriptionally splice the sequence. See here for more details:

  • $\begingroup$ why does it transcribe 2.4 million bases and cut it down to 14,000? Seems like a waste of time and energy, not to mention the potential for mutations, which if muscular dystrophy is any clue, would be deadly. $\endgroup$
    – user137
    Sep 15 '14 at 22:26
  • $\begingroup$ alternative splicing is a wierd thing to be sure. biology is not rational but then human science has not yet produced a living cell from first principles... $\endgroup$
    – shigeta
    Sep 15 '14 at 22:28
  • $\begingroup$ Cases like this make me wonder why alternative splicing would be worth it. If you divide 2.4 million bases by 14000 bases you get 171.4. Seems like you could just make multiple copies of dystrophin to cover all the isoforms and still save space, and only need to transcribe the copy you want. Evolution should have gotten an engineering degree. $\endgroup$
    – user137
    Sep 15 '14 at 23:06
  • $\begingroup$ I don't have a reference for this, but I was told by a professor once that a long developmental gene is Drosophila takes longer to transcribe than the length of the cell cycle during initial development so that is not actually expressed until the cell cycle lengthens later in development. That's perhaps one answer (if uncommon) to the question of why some genes are so long. $\endgroup$
    – canadianer
    Sep 16 '14 at 0:30

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.