Note that the following awk code produces both output files you want in one invocation (with pairs with 2nd lines longer than 15 characters in a file name long and pairs with 2nd lines shorter than 16 characters in a file named short):
If you want to try this on a Solaris/SunOS system, change awk to nawk or /usr/xpg4/bin/awk.
These 3 Users Gave Thanks to Don Cragun For This Post:
Hi ,
I have a peculiar case, where my sed command is working on a file which contains lines of small length.
sed "s/XYZ:1/XYZ:3/g" abc.txt > xyz.txt
when abc.txt contains lines of small length(currently around 80 chars) , this sed command is working fine.
when abc.txt contains lines of... (3 Replies)
My files look like this
And I need to cut the sequences at the last "A" found in the following 'pattern' -highlighted for easier identification, the pattern is the actual file is not highlighted.
The expected result should look like this
Thus, all the sequences would end with AGCCCTA... (2 Replies)
This is what I would like to accomplish, I have an input file (file A) that consist of thousands of sequence elements with the same number of characters (length), each headed by a free text header starting with the chevron ‘>' character followed by the ID (all different IDs with different lenghts)... (9 Replies)
My file looks something like this
Wnat I need is to look for the Reference sequence (">Reference1") and based on the length of that sequence trim all the entries in that file. So, the rersulting file will contain all sequences with the same length, like this
Thus, all sequences will keep... (5 Replies)
Hi,
I have a file with more than 28000 records and it looks like below..
>mm10_refflat_ABCD range=chr1:1234567-2345678
tgtgcacactacacatgactagtacatgactagac....so on
>mm10_refflat_BCD range=chr1:3234567-4545678...
tgtgcacactacacatgactagtatgtgcacactacacatgactagta
.
.
.
.
.
so on
... (2 Replies)
I have a fastq file from small RNA sequencing with sequence lengths between 15 - 30. I wanted to filter sequence lengths between 21-25 and write to another fastq file. how can i do that? (4 Replies)
I have two files with thousands of sequences of different lengths. infile1 contains the actual sequences and infile2 the scores for each A, T, G and C in infile1. Something like this:
infile1:
>HZVJKYI01ECH5R
TTGATGTGCCAGCTGCCGTTGGTGTGCCAA
>HZVJKYI01AQWJ8
GGATATGATGATGAACTGGTTTGGCACACC... (4 Replies)
I have to remove sequences from a file based on the distance value. I am attaching the file containing the distances (Distance.xls)
The second file looks something like this:
Sequences.txt
>Sample1 Freq 59
ggatatgatgatgaactggt
>Sample1 Freq 54
ggatatgatgttgaactggt
>Sample1 Freq 44... (2 Replies)
I have a list of IDs in file1 and a list of sequences in file2. I can print sequences from file2, but I'm asking for help in printing the sequences in the same order as the IDs appear in file1.
file1:
EN_comp12952_c0_seq3:367-1668
ES_comp17168_c1_seq6:1-864
EN_comp13395_c3_seq14:231-1088... (5 Replies)
I could calculate the length of entire fasta sequences by following command,
awk '/^>/{if (l!="") print l; print; l=0; next}{l+=length($0)}END{print l}' unique.fasta
But, I need to calculate the length of a particular fasta sequence specified/listed in another txt file. The results to to be... (14 Replies)
Discussion started by: dineshkumarsrk
14 Replies
LEARN ABOUT OSX
ucblinks
ucblinks(1B) SunOS/BSD Compatibility Package Commands ucblinks(1B)NAME
ucblinks - adds /dev entries to give SunOS 4.x compatible names to SunOS 5.x devices
SYNOPSIS
/usr/ucb/ucblinks [-e rulebase] [-r rootdir]
DESCRIPTION
ucblinks creates symbolic links under the /dev directory for devices whose SunOS 5.x names differ from their SunOS 4.x names. Where possi-
ble, these symbolic links point to the device's SunOS 5.x name rather than to the actual /devices entry.
ucblinks does not remove unneeded compatibility links; these must be removed by hand.
ucblinks should be called each time the system is reconfiguration-booted, after any new SunOS 5.x links that are needed have been created,
since the reconfiguration may have resulted in more compatibility names being needed.
In releases prior to SunOS 5.4, ucblinks used a nawk rule-base to construct the SunOS 4.x compatible names. ucblinks no longer uses nawk
for the default operation, although nawk rule-bases can still be specifed with the -e option. The nawk rule-base equivalent to the SunOS
5.4 default operation can be found in /usr/ucblib/ucblinks.awk.
OPTIONS -e rulebase Specify rulebase as the file containing nawk(1) pattern-action statements.
-r rootdir Specify rootdir as the directory under which dev and devices will be found, rather than the standard root directory /.
FILES
/usr/ucblib/ucblinks.awk sample rule-base for compatibility links
ATTRIBUTES
See attributes(5) for descriptions of the following attributes:
+-----------------------------+-----------------------------+
| ATTRIBUTE TYPE | ATTRIBUTE VALUE |
+-----------------------------+-----------------------------+
|Availability |SUNWscpu |
+-----------------------------+-----------------------------+
SEE ALSO devlinks(1M), disks(1M), ports(1M), tapes(1M), attributes(5)SunOS 5.10 13 Apr 1994 ucblinks(1B)