Getting mRNA coding sequence for a UCSC accession

Question

[jacobsen.jeremy]

I cannot figure out how to pull out the coding sequence from KnownGeneMrna.

I know that sequences in KnownGeneMrna contain UTRs, so what I am doing is taking CDS_start - tx_start from KnownGene to find the start of the CDS from the beginning of the KnownGeneMrna sequence. The columns in KnownGene are:

{accession},{chrom},{strand},{tx_start},{tx_end},{CDS_start},{CDS_end} etc..

The problem is that some transcripts are shorter than the offset! For instance uc010nyq.2. Why is this and what am I doing wrong? I have found other related posts but none that address this point.

Thanks,
Jeremy

[Pierre Lindenbaum]

The problem is that some transcripts are shorter than the offset!

How did you find this?

$ mysql --user=genome --host=genome-mysql.cse.ucsc.edu -A -D hg38 -e 'select count(*) from knownGene where cdsStart<txStart'
+----------+
| count(*) |
+----------+
|        0 |
+----------+

For instance uc010nyq.2

$ mysql --user=genome --host=genome-mysql.cse.ucsc.edu -A -D hg38 -e 'select * from knownGene where name="uc010nyq.2"\G'
*************************** 1. row ***************************
      name: uc010nyq.2
     chrom: chr1
    strand: +
   txStart: 1616309
     txEnd: 1630610
  cdsStart: 1623410
    cdsEnd: 1630530
exonCount: 19
exonStarts: 1616309,1623388,1623773,1624794,1624990,1625285,1625545,1626649,1626836,1627073,1627295,1627672,1628018,1628272,1628488,1629132,1629384,1629638,1630291,
  exonEnds: 1616614,1623699,1623945,1624901,1625185,1625428,1625653,1626754,1626999,1627207,1627444,1627829,1628179,1628399,1628722,1629311,1629566,1629704,1630610,
proteinID: Q96AX9
   alignID: uc010nyq.2

Where is the problem?

[jacobsen.jeremy]

The problem is that I'm misinterpreting something; probably to do with the contents of knownGeneMrna.

uc010nyq.2    chr1    +    1551689    1565990    1558790    1565910

1558790-1551689 = 7101

I take this to mean (erroneously I'm sure) that if I pull the sequence from knownGeneMrna then the coding sequence will begin 7101 bps from the start of the transcript.

The tx length for uc010nyq.2 from knownGeneMrna is 3317, so this cannot be the right assumption.

[jacobsen.jeremy]

Thanks Pierre. I could do it this way if I were pulling the coding sequences directly from the hg19 chromosome fasta files. This seems silly though because they are already assembled in knownGeneMrna. The problem is that the UTRs are included in these sequences and I want to remove the UTRs so as to be left with the coding sequence.

Below is the unabridged sequence for uc010nyq.2 taken from knownGeneMrna. The section bounded by [ (line 5) and ] (the penultimate line) is the coding sequence. The UTR offset to the start of the coding sequence is 327 characters. Where is this offset annotated or is there another file that only has the CDS?

cgcaattacgggcccggcgctggcggctcctgcgcgctcagaccccagggagcccatccgggcaggcggcggccctgagt
gtcgcggccgtgggcccgagtggacctggagccggcgggcagccccgggggcagacaggcgaccgagccgcgggtcgagg
tgctaactgtgcatcttggcatctcccctcggccacagggttggaagcccagcgaggctagaggccagtcccaaagtttc
caggcatcagggctgcagcccaggagcctcaaggcggcccggcgggcgactggacggccggacaggtcccgagcagcccg
gcccacc[atggacccctctgcccacaggtcccgagcagccccgcccaacatggacccagacccccaggcgggcgtgcag
gtgggcatgcgggtggtgcgcggcgtggactggaagtggggccagcaggacggcggcgagggcggcgtgggcacggtggt
ggagcttggccgccacggcagcccctcgacacccgaccgcacagtggtcgtgcagtgggaccagggcacgcgcaccaact
accgcgccggctaccagggcgcgcacgacctgctgctgtacgacaacgcccagatcggcgtccggcaccccaacatcatc
tgtgactgctgcaagaagcacgggctgcgggggatgcgctggaagtgccgtgtgtgcctggactacgacctctgcacgca
gtgctacatgcacaacaagcatgagctcgcccacgccttcgaccgctacgagaccgctcactcgcgccctgtcacactga
gtccccgccagggcctcccgaggatcccactaaggggcatcttccagggagcgaaggtggtgcgaggccccgactgggag
tggggctcacaggatggaggggaagggaaaccgggccgtgtggtggacatccgtggctgggatgtggagacaggccggag
tgtggccagcgtgacgtgggctgatggtaccaccaatgtgtaccgtgtgggccacaagggcaaggtggacctcaagtgtg
tgggcgaggcagcgggcggcttctactacaaggaccacctcccaaggctcggcaagccggcggagctgcagcgcagggtg
agtgctgacagccagcccttccagcacggggacaaggtcaagtgtctgctggacactgatgtcctgcgggagatgcagga
aggccacggcggctggaaccccaggatggcggagtttatcggacagacgggcaccgtgcatcgtatcacggaccgcgggg
acgtgcgcgtgcagttcaaccacgagacgcgctggaccttccaccccggggcgctcaccaagcaccactccttctgggtg
ggcgacgtggtccgggtcatcggcgaccttgacacagtgaagcggctgcaggctgggcatggcgagtggacggacgacat
ggcccctgccctgggccgcgtcgggaaggtggtgaaagtgtttggagacgggaacctgcgtgtagcagtcgctggtcagc
ggtggaccttcagcccctcctgcctggtggcctaccggcccgaggaggatgccaacctggacgtggccgagcgcgcccgg
gagaacaaaagctcactgagcgtggccctggacaagcttcgggcccagaagagtgacccagagcacccgggaaggctggt
ggtggaggtggcgctgggtaacgcagcccgggctctggacctgctgcggaggcgcccagagcaggtggacaccaagaacc
aaggcaggaccgctctgcaagtggctgcctacctgggccaggtggagttgatacggctgctgctacaagccagggcgggc
gtggacctgccggacgacgagggcaacacggcactgcactacgcggccctggggaaccagcccgaggccaccagggtgct
cctgagtgctgggtgccgggcggacgccatcaacagcacccagagcacagcactgcacgtggccgtgcagaggggcttcc
tggaggtggtgcgggccctgtgtgagcgcggctgtgacgtcaacctgcccgacgcccactcggacacgcccctgcactcc
gccatctcggcgggcactggagccagcggcattgtcgaggtcctcacggaggtgccaaacatcgatgttaccgccaccaa
cagccagggtttcaccctgctgcaccatgcctccctcaagggtcacgcgctagctgtgagaaagattctggctcgggcgc
ggcagctggtggacgccaagaaggaggacggcttcacggcgctgcatctggctgccctcaacaaccaccgcgaggtggcc
cagatcctcatccgggagggccgctgtgacgtgaacgtgcgcaaccggaagctgcagtccccgctgcatctcgccgtgca
acaggcccacgtggggctggtgccgctactggtggacgctgggtgcagtgtcaacgccgaggacgaggagggggacacag
ccctgcacgtggcgctgcagcgtcatcagctgctgcccctggtggctgatggggccgggggggacccagggcccttgcag
ctgctgtccaggctacaggcctcgggcctccccggcagcgcggagctgacggtgggcgcggcggtcgcctgcttcctggc
gctggagggcgccgacgtgagctacaccaaccaccgcggtcggagcccgctggacctggccgccgagggtcgcgtgctca
aggcccttcagggctgcgcccagcgcttccgggagcggcaggcgggcgggggcgcggccccgggccccaggcaaacgctc
gggacccccaacaccgtgacgaacctgcacgtgggcgccgcgccggggcccgaggccgctgagtgcctggtgtgctccga
gctggcgctgctggtgctgttctcgccgtgccagcaccgcaccgtgtgtgaggagtgcgcgcgcaggatgaagaagtgca
tcaggtgccaggtggtcgtcagcaagaaactgcgcccagacggctctgaggtggcgagcgccgcccccgcccccggcccg
ccgcgccagctggtggaggagctgcagagccgctaccggcagatggaggaacgcatcacctgccccatctgcatcgacag
ccacatccgcctcgtgttccagtgcggccacggcgcatgcgccccctgcggctccgcgctcagcgcctgccccatctgcc
gccagcccatccgcgaccgcatccagatcttcgtgtga]gccgcgccgtccgccgcgcccgagctgccttcgcgtgcccc
cgccctgtgttttataaaaagaaagattctcggacgttg

[Pierre Lindenbaum]

Please don't post your comments as a new answer.

Answer 1

[Pierre Lindenbaum]

You cannot just subtract cdsStart - txStart because there is one the two positions are localized one two distinct exons:

• exon1: 1616309-1616614
• intron1: 1616614-1623388
• exon2:1623388-1623699

 txStart: 1616309
 cdsStart: 1623410

[jacobsen.jeremy]

Thanks Pierre. I could do it this way if I were pulling the coding sequences directly from the hg19 chromosome fasta files. This seems silly though because they are already assembled in knownGeneMrna. The problem is that the UTRs are included in these sequences and I want to remove the UTRs so as to be left with the coding sequence.

Below is the unabridged sequence for uc010nyq.2 taken from knownGeneMrna. The section in yellow is the coding sequence. The UTR offset to the start of the coding sequence is 327 characters. Where is this offset annotated or is there another file that only has the CDS?

cgcaattacgggcccggcgctggcggctcctgcgcgctcagaccccagggagcccatccgggcaggcggcggccctgagtgtcgcggccgtgggcccgagtggacctggagccggcgggcagccccgggggcagacaggcgaccgagccgcgggtcgaggtgctaactgtgcatcttggcatctcccctcggccacagggttggaagcccagcgaggctagaggccagtcccaaagtttccaggcatcagggctgcagcccaggagcctcaaggcggcccggcgggcgactggacggccggacaggtcccgagcagcccggcccacc***atggacccctctgcccacaggtcccgagcagccccgcccaacatggacccagacccccaggcgggcgtgcaggtgggcatgcgggtggtgcgcggcgtggactggaagtggggccagcaggacggcggcgagggcggcgtgggcacggtggtggagcttggccgccacggcagcccctcgacacccgaccgcacagtggtcgtgcagtgggaccagggcacgcgcaccaactaccgcgccggctaccagggcgcgcacgacctgctgctgtacgacaacgcccagatcggcgtccggcaccccaacatcatctgtgactgctgcaagaagcacgggctgcgggggatgcgctggaagtgccgtgtgtgcctggactacgacctctgcacgcagtgctacatgcacaacaagcatgagctcgcccacgccttcgaccgctacgagaccgctcactcgcgccctgtcacactgagtccccgccagggcctcccgaggatcccactaaggggcatcttccagggagcgaaggtggtgcgaggccccgactgggagtggggctcacaggatggaggggaagggaaaccgggccgtgtggtggacatccgtggctgggatgtggagacaggccggagtgtggccagcgtgacgtgggctgatggtaccaccaatgtgtaccgtgtgggccacaagggcaaggtggacctcaagtgtgtgggcgaggcagcgggcggcttctactacaaggaccacctcccaaggctcggcaagccggcggagctgcagcgcagggtgagtgctgacagccagcccttccagcacggggacaaggtcaagtgtctgctggacactgatgtcctgcgggagatgcaggaaggccacggcggctggaaccccaggatggcggagtttatcggacagacgggcaccgtgcatcgtatcacggaccgcggggacgtgcgcgtgcagttcaaccacgagacgcgctggaccttccaccccggggcgctcaccaagcaccactccttctgggtgggcgacgtggtccgggtcatcggcgaccttgacacagtgaagcggctgcaggctgggcatggcgagtggacggacgacatggcccctgccctgggccgcgtcgggaaggtggtgaaagtgtttggagacgggaacctgcgtgtagcagtcgctggtcagcggtggaccttcagcccctcctgcctggtggcctaccggcccgaggaggatgccaacctggacgtggccgagcgcgcccgggagaacaaaagctcactgagcgtggccctggacaagcttcgggcccagaagagtgacccagagcacccgggaaggctggtggtggaggtggcgctgggtaacgcagcccgggctctggacctgctgcggaggcgcccagagcaggtggacaccaagaaccaaggcaggaccgctctgcaagtggctgcctacctgggccaggtggagttgatacggctgctgctacaagccagggcgggcgtggacctgccggacgacgagggcaacacggcactgcactacgcggccctggggaaccagcccgaggccaccagggtgctcctgagtgctgggtgccgggcggacgccatcaacagcacccagagcacagcactgcacgtggccgtgcagaggggcttcctggaggtggtgcgggccctgtgtgagcgcggctgtgacgtcaacctgcccgacgcccactcggacacgcccctgcactccgccatctcggcgggcactggagccagcggcattgtcgaggtcctcacggaggtgccaaacatcgatgttaccgccaccaacagccagggtttcaccctgctgcaccatgcctccctcaagggtcacgcgctagctgtgagaaagattctggctcgggcgcggcagctggtggacgccaagaaggaggacggcttcacggcgctgcatctggctgccctcaacaaccaccgcgaggtggcccagatcctcatccgggagggccgctgtgacgtgaacgtgcgcaaccggaagctgcagtccccgctgcatctcgccgtgcaacaggcccacgtggggctggtgccgctactggtggacgctgggtgcagtgtcaacgccgaggacgaggagggggacacagccctgcacgtggcgctgcagcgtcatcagctgctgcccctggtggctgatggggccgggggggacccagggcccttgcagctgctgtccaggctacaggcctcgggcctccccggcagcgcggagctgacggtgggcgcggcggtcgcctgcttcctggcgctggagggcgccgacgtgagctacaccaaccaccgcggtcggagcccgctggacctggccgccgagggtcgcgtgctcaaggcccttcagggctgcgcccagcgcttccgggagcggcaggcgggcgggggcgcggccccgggccccaggcaaacgctcgggacccccaacaccgtgacgaacctgcacgtgggcgccgcgccggggcccgaggccgctgagtgcctggtgtgctccgagctggcgctgctggtgctgttctcgccgtgccagcaccgcaccgtgtgtgaggagtgcgcgcgcaggatgaagaagtgcatcaggtgccaggtggtcgtcagcaagaaactgcgcccagacggctctgaggtggcgagcgccgcccccgcccccggcccgccgcgccagctggtggaggagctgcagagccgctaccggcagatggaggaacgcatcacctgccccatctgcatcgacagccacatccgcctcgtgttccagtgcggccacggcgcatgcgccccctgcggctccgcgctcagcgcctgccccatctgccgccagcccatccgcgaccgcatccagatcttcgtgtga***gccgcgccgtccgccgcgcccgagctgccttcgcgtgcccccgccctgtgttttataaaaagaaagattctcggacgttg

[jacobsen.jeremy]

Got it Pierre. Thanks!

cdsStart-txStart-intron1