t*****z
发帖数: 1598 | 1 amplicon sequencing，最简单最传统的二代测序。只要设计好针对该基因的引物就可
以测了。通过multiplexing，测序仪里一个洞可以塞很多样品。具体怎么操作问测序中
心就可以。 |
|
n******7
发帖数: 12463 | 2 我说的是sequencer方面的问题
一个library片段太短的话很容易让其他library的reads少很多
上样的时候不好估计
后续处理问题应该不大
因为每个library的size distribution都会单独测定
而sequencing reads demultiplex之后也没啥影响 |
|
B*********h
发帖数: 800 | 3 ☆─────────────────────────────────────☆
BulletTooth (DesertEagle.50) 于 (Sun Jul 9 21:59:14 2006) 提到:
Character Sequence Monitoring
面试公司：
面试部门：
面试职位：
*****
问题：
*****
Unless you have been asked to use a specific language write the program in any
programming language that you know well (no pseudo-code). The program
should monitor a possibly infinite stream of characters from the keyboard (
standard input). If it detects the sequence "aaa" it outputs a "0". If
it detects the seq |
|
m****r
发帖数: 141 | 4 The interview has been done.
Give a sequence of data (with duplicates), move a fix-sized window along the
data sequence and find mode in the window at each iteraion, where the
oldest data is removed and a new data is inserted to the window.
I cannot find better solutions here.
My idea: Use a hashtable, key is the data, key's data is the frequency of
the data occuring in the window.
At the first iteration, iterate each data in the window and put it to the
hashtable, meanwhile cout the frequency o... 阅读全帖 |
|
a****o
发帖数: 1786 | 5 【 以下文字转载自 Biology 讨论区 】
发信人: cnDiehard (半杯), 信区: Biology
标 题: 也谈next-generation sequencing technology
发信站: BBS 未名空间站 (Sun Nov 29 14:41:44 2009, 美东)
这几天看了toptip的高通量DNA测序技术在生物学研究中的应用,,受益匪浅.. 俺们实验
室的研究也focus这种应用..所以平时也作了很多数据分析.. 这里比较一下两种测序的
平台(solid和solexa),,相关的软件,,也讨论一下他们的应用.. 本人经验有限,,本文权
当抛砖引玉,,希望各位大侠多多指教:
1. raw sequence 编码
SOLiD: Solid采用color-coded序列输出. 所以输出的序列除了首位以外,,都是数字(0-
3). 没两位编码一个nucleotide.. T0132123. 然后根据一个编码表,,可以转换成
nucleotide序列.
Solexa: 输出采用nucleotide编码, 不需要任何转换..
2. yield
两种测序技术每个单 |
|
g**y
发帖数: 366 | 6 最近开始准备攒审稿，广发邮件几十封，有零星几个回复说把我加到reviewer list，
但是两个月了一篇也没收到，真急阿。
恳请这里的好心人若有合适的稿件考虑一下我，或者请帮我推荐一下，非常感谢！
我的研究领域是：computational biology, gene regulation, epigenetics, Next-
generation sequencing, microarray data analysis, machine learning, data
mining, clustering/classification etc.
自己发表过几篇computational biology, algorithms 等方面的paper, 最高impact
factor 8.02
若需要详细个人信息、publication list的话，请站内联系，谢谢！ |
|
b********h
发帖数: 119 | 7 subsequence不是subarray吧。bruteforce应该是2^n。
DP的做法跟max subsequence一样，定义S(i)为在i结束的max increaseing sequence，
则，
S(i) = max{S(j)+A[i], A[i]>A[j] && 0<=j |
|
|
y**i
发帖数: 1112 | 9 我想的就是这样的，两个循环，外层循环是当前检测的subsequence的起始点，内层循
环是终结点，用一个变量max记录目前为止最大的sum，用另一个变量index记录上一个
检测的increasing subsequence的终结点，如果当前的内循环变量所指的元素大于变量
index所指元素，那么就更新increasing subsequence的终结点并把其于max的和跟max
比较及更新max。对这应该是DP思想，因为记录了检测过的increasing sequence的最大
和，而不是把所以的increasing subsequence先全部找出来再求和比较，那样确实是2^
n。
不过这个DP应该不需要额外O(n)空间，因为我们只要记录最大就可以了对吧，没必要记录所有的sum |
|
Z*****Z
发帖数: 723 | 10 问题是如果内循环变量所指元素小于index所指元素应该如何做呢？
我想的就是这样的，两个循环，外层循环是当前检测的subsequence的起始点，内层循
环是终结点，用一个变量max记录目前为止最大的sum，用另一个变量index记录上一个
检测的increasing subsequence的终结点，如果当前的内循环变量所指的元素大于变量
index所指元素，那么就更新increasing subsequence的终结点并把其于max的和跟max
比较及更新max。对这应该是DP思想，因为记录了检测过的increasing sequence的最大
和，而不是把所以的increasing subsequence先全部找出来再求和比较，那样确实是2^
n。
不过这个DP应该不需要额外O(n)空间，因为我们只要记录最大就可以了对吧，没必要记
录所有的sum |
|
Z*****Z
发帖数: 723 | 11 想了一下，觉得玄。
LIS优化的巧妙之处在于sequence长度不会超过输入数组的长度，对每个increasing su
bsequence可以根据其长度记录那个最后的元素。
这里是和，貌似不太可能 |
|
C***y
发帖数: 2546 | 12 在 inorder sequence中找root node怎么搜索比较快？
我的想法是从中间，同时向两边linear search
还有什么更好的办法吗？ |
|
C***y
发帖数: 2546 | 13 一般情况下，root应该在inorder sequence中比较接近中间的地方 |
|
c*******r
发帖数: 309 | 14 Given an array of random numbers. Find the longest consecutive sequence.
For ex
Array 100 3 200 1 2 4
Ans 1 2 3 4
Array 101 2 3 104 5 103 9 102
Ans 101 102 103 104
Can we do it in one go on array using extra space??
我想到的方式是先sort， 然后用index来存,然后记录max长度. 不断更新新的index指
针.
不过据说复杂度要控制在o(n).
没什么好的想法啊 |
|
x******a
发帖数: 6336 | 15 例如ABABCD, BABDC,
longest common consecutive sequence 是“BAB”
谢谢 |
|
p*****2
发帖数: 21240 | 16 像是DP，但是没太明白题。
是找两个字符串里的common consecutive sequence? 为什么ABA会是一个解呢？ |
|
V******J
发帖数: 9 | 17 正算倒算都可以。正算:dp[i][j]表示已(i,j)结束的最长sequence. |
|
a***e
发帖数: 413 | 18 自己做的总是memory exceeded，
看了讨论，
https://oj.leetcode.com/discuss/24478/i-did-it-in-10-lines-of-c
还是不懂为啥m[t = t << 3 & 0x3FFFFFFF | s[i] & 7]++ == 1就能达到rolling hash
的效果呢?
i只是一个char啊，这个是哪里推导出来的？多谢！
Neat idea. The additional 1 bit per letter still encode each substring in
10x3 = 30 bits, just under 4 bytes for a 32-bit integer.
Your code could be further simplified. By observing that s[i] & 7 is never 0
, each of the first nine substrings with length < 10 will have unique hash
key and will never collid... 阅读全帖 |
|
x******8
发帖数: 48 | 19 Longest Increasing Sequence 2D matrix 二维数组最长递增子序列，只能上下左右，
不能对角线，找出长度或者移动路径，这个怎么做？brutal force? |
|
t******e
发帖数: 37 | 20 There are two job openings for research assistant. The lab is a core lab in
a hospital medical center with focus on DNA sequencing,NGS etal. the salary
is around $50,000 depending on the background and experience. Good benefit
and stable position. If interested, pls forward your CV. Thanks |
|
b*p
发帖数: 242 | 21 Commercializing Full Genome Sequencing (Genome Bubble 2.0?)
While service is the future (GNOM), currently it is still about equipment R&
D (PACB), though.
Any Comments? |
|
w*****u
发帖数: 4768 | 22 please have a look at ILMN, it has the most advanced sequencing machine |
|
x*********a
发帖数: 906 | 23 .edu email
Ph.D degree
has background on sequencing
Thank you |
|
x*********a
发帖数: 906 | 24 .edu email
Ph.D degree
has background on sequencing
Thank you |
|
A*F
发帖数: 2272 | 25 有两篇manuscript需要reviewers
领域是SNP analysis和exome sequencing
请感兴趣的同学发简历或近期文章列表至站内信或[email protected]/* */
谢谢
需要一定bioinfo基础，只做wet lab的同学可能不行 |
|
a****0
发帖数: 60 | 26 n久不打了。准备国庆节骚扰他们。
what's the sequence to reach a real person? 3x |
|
y****m
发帖数: 1322 | 27 人笨，请教大家一个问题，谢谢啦。
choose the anwser that best continues the sequence.
ΔΟΟΟ ΔΟΔΟ ΔΟΔΔ
a) ΔΟΟΟ b)ΔΟΔΟ c) ΟΔΟΔ |
|
c****u
发帖数: 3277 | 28 if you play 1NT as forcing or semiforcing, it's usually not a good
idea to introduce your 4 card spade suit with about 16-bad 18 HCP,
because partner usually denies spade suit. so you should normally rebid
2C or 2D here. Thus, we have a free bid: 2S. This 2S can be treated as
GF and starts a relay sequence.
1H-1N-2S:
2NT:relay.
3C/D: natural.
3H: 6 card hearts, about 19 HCP, ask partner to place the final contract.
3S: 5-6 in majors.
2NT: 4-5 in majors.
4C/D/H: self splinter in C/D/S, |
|
c****u
发帖数: 3277 | 29 within this framework, we now may have a partnership agreement
on how to stop at 4 level minor.
When both sides show minimum, 4 level minor is passable:
1S 2D
2S 2NT(relay)
3D 4D(invitational)
1S 2D
2S 3D
4D(forcing! because 3D sets up gameforcing sequence)
1S 2D
2S 2N(relay)
3C 4C(invitational)
1S 2D
2S 3C
4C(forcing! because 3C shows extra)
1S 2D
2H 2N(relay)
3D 4D(invitational)
1S 2D
2H 2N(relay)
3C 4C(forcing! 4th suit by declarer also shows extra)
1S 2D
2S 2N
3D 3S
4D (invitational, because |
|
v*****r
发帖数: 2325 | 30 modified my swing recently, from 2-plane to 1-plane, trying to learn 1-plane.
1-plane ask more using core muscle and more commitment to backswing.
the main stand of 1-plane is to reduce variable factors and make it simple.
for more check out:
http://www.jeffrittergolf.com/wiki/show/Jeff%20Ritter%20Golf%20-%20Home
not yet hit balls at range yet, but feel that hit better using 1-plane.
the sequence taken from slo mo video from DLT and face-on, using practice
ball in backyard.
only one week learnin |
|
c*****c
发帖数: 30 | 31 What software did you use to capture the swing sequence?
Thanks. |
|
|
|
R*s
发帖数: 2041 | 34 【 以下文字转载自 Biology 讨论区 】
发信人: biospiner (biospiner), 信区: Biology
标 题: Re: 请问protein sequence alignment怎么做？
发信站: BBS 未名空间站 (Sun Nov 5 00:16:09 2006)
http://www.ebi.ac.uk/clustalw/index.html |
|
c**t
发帖数: 2744 | 35 Generated a sequence diagram for a public web service, the graph is too
complex to understand/draw/review. I feel more comfortable by reading code
rather than the graph. |
|
z*****7
发帖数: 367 | 36 请问win10 怎么改 boot driver sequence， 系统migrate到ssd, 进bios后，boot选项
，只看到一个boot manager，而且看不到boot manager里面怎么设置的。
请问大侠怎么改成ssd启动啊。谢谢 |
|
m******t
发帖数: 2416 | 37
Not sure I understand what you mean by "sequence不连着了"?
Also what database? |
|
b**********g
发帖数: 806 | 38 刚刚在dev的环境下试了一下,估计production上是有人从database里手动删行了,所以那
个sequence就不work了 |
|
i**********r
发帖数: 36 | 39 【 以下文字转载自 Linux 讨论区 】
【 原文由 interstellar 所发表 】
i have a sequence of postscript files and want to slide show them with
a 0.5 second pause between files. i tried animate, which is too slow;
i also tried gs, which is very fast, but no pause can be set between
two files. are there any other programs in linux to do this? |
|
g****r
发帖数: 14 | 40 it is possible the gene in the bank has mistakes
but what pcr enzyme did you use, since taq tends to give errors
besides, you may order the clone from IMAG directly, instead of cloning
by yourself, if you are sure the clone is the one that you are going to use
I guess it is cheaper to buy the clone, than do the whole cloning/sequencing by your own |
|
m******n
发帖数: 194 | 41 go to genome.ucsc.edu, if you have partial sequence of the gene, use
blat to search the genome, you will have a genome coordinate on the top
of the output graph, exact that piece of genomic DNA |
|
c*******d
发帖数: 192 | 42 这几天看了toptip的高通量DNA测序技术在生物学研究中的应用,,受益匪浅.. 俺们实验
室的研究也focus这种应用..所以平时也作了很多数据分析.. 这里比较一下两种测序的
平台(solid和solexa),,相关的软件,,也讨论一下他们的应用.. 本人经验有限,,本文权
当抛砖引玉,,希望各位大侠多多指教:
1. raw sequence 编码
SOLiD: Solid采用color-coded序列输出. 所以输出的序列除了首位以外,,都是数字(0-
3). 没两位编码一个nucleotide.. T0132123. 然后根据一个编码表,,可以转换成
nucleotide序列.
Solexa: 输出采用nucleotide编码, 不需要任何转换..
2. yield
两种测序技术每个单位的yield都在15M左右 (最近一次solexa平均yield约20M per
Lane).. 好像solid是spot,, solexa叫lane..solid每次可以测8个spot,, solexa每次
可以测序8个lane.. 所以从raw输出的数量来说,,两个没有大的差别..
3. S |
|
A**H
发帖数: 4797 | 43 谢谢分析
今年早些时候Genome Biology有一篇文章做了三家的比较，大致的结论是Illumina得最
好。观点是否有失偏颇尚待其他研究佐证。
Evaluation of NGS platforms for population targeted sequencing studies
我认为SOLiD的问题在它的color-space上面：它最终要死在它的这个color-code上面。
它这个color-code让搞生物的人很糊涂，甚至让搞数据分析的人也难把握其中奥妙。比
如说楼上所说的“b. solid用所有35bp作alignment,,最多允许6个mismatches”，我的
理解肯定是6个color space mismatches，而不是nucleotide space mismatches，大概
地说，一个nucleotide space mismatch 小于或这等于 2个 color space mismatches
。当然即便如此，允许6个color space mismatches in 35-bp 还是太多了。
SOLiD的技术小组已经注意到了这个问题，但是 |
|
H*****e
发帖数: 120 | 44 RNA sequence technology is now becoming more and more frequent in seminar.
What is the advantage and
disadvantage? Can people who has experience make some comments? |
|
j*****q
发帖数: 82 | 45 rt
how to amplify extreme AT-rich sequences on genomic DNA?
followed some literatures to decrease the enlongation temp. to 60 degree,
but still failed.
i am using the roche polymerase and qiagen one.
seeking for helps!
please!!! |
|
H****N
发帖数: 997 | 46 Sequence ChIP-seq samples. Do you have experience collaborating with them?
Thanks. |
|
s*********e
发帖数: 399 | 47 没有collaboration, 不过据说他们要开美国分部.
虽然版上不少人鄙视他们，不过做sequence,恐怕他们还有实力吧。 最近两个月已经两
篇nature 封面了。 而且其他的几个什么古人阿之类的封面，其实也是他们做的。
如果，你有兴趣，我可以让朋友问问价钱。 |
|
p**********t
发帖数: 2636 | 48 2 people in my lab use Staden package. It's not very user-friendly but it's
free and does its job.
The rest people use Sequencher, which you need a key and have to wait in
line to use it. A postdoc who tried many sequence analysis programs told me
that so far Sequencher is the best, but you need to persuade your advisor to
buy it. |
|
f**********e
发帖数: 1994 | 49 如果 color space sequence 能 map 上 reference, SOLiD 理论上
和实际上精确度都会比较好。SOLiD 的 encoding 实际上就是 EE 用的
纠错码。 |
|
|
