ScienceNow: Hunting Down Cancer

Cancer researchers have long sought an alternative treatment to chemotherapy that can destroy tumors without damaging the rest of the body. A new study with mouse and human tumors may put such a treatment within reach.

Designing drugs to target cancer is a challenge because cancerous cells are well camouflaged within otherwise healthy tissues. But researchers have begun to find molecular tags unique to cancer cells. Last year, Wadih Arap and Renata Pasqualini, a husband-and-wife team of cancer biologists at the M. D. Anderson Cancer Center at the University of Texas, Houston, identified a protein called GRP78 expressed by prostate cancer cells and not by healthy tissues. They knew they'd found a promising drug target because GRP78 is expressed on the cell surface, where it's easily accessible to drugs. Additionally, a drug targeted to GRP78 should be unlikely to attack healthy cells because GRP78 is a so-called stress response protein which is only expressed on the surface of cells under stressful conditions such as those that occur within the oxygen-poor lump of a tumor.

To test GRP78 as a target, a team led by Arap and Pasqualini designed a short string of amino acids that binds specifically to the surface of the GRP78 protein. They then fused this string to a small corkscrew-shaped protein that, when internalized, triggers the cell to commit suicide. The team hoped that the hybrid molecule would act as a tumor-seeking missile when injected into the bloodstream.

The missile appears to be deadly accurate. Reporting in the September issue of Cancer Cell, the team shows that its hybrid molecule located and destroyed human prostate tumors transplanted into mice, without harming any other kind of tissue. It did the same for breast tumors. The next step, say Arap and Pasqualini, is a series of preclinical studies to determine its safety for clinical trials in humans.

Over the past decade, Arap and Pasqualini have pioneered the "smart bomb" approach to cancer therapy, says Bruce Zetter, a cancer researcher at Harvard Medical School in Boston. This study "shows how close we are getting" to putting the strategy into action.

--JOHN BOHANNON

 
基于python实现的粒子群的VRP(车辆配送路径规划)问题建模求解+源码+项目文档+算法解析,适合毕业设计、课程设计、项目开发。项目源码已经过严格测试,可以放心参考并在此基础上延申使用,详情见md文档 算法设计的关键在于如何向表现较好的个体学习,标准粒子群算法引入惯性因子w、自我认知因子c1、社会认知因子c2分别作为自身、当代最优解和历史最优解的权重,指导粒子速度和位置的更新,这在求解函数极值问题时比较容易实现,而在VRP问题上,速度位置的更新则难以直接采用加权的方式进行,一个常见的方法是采用基于遗传算法交叉算子的混合型粒子群算法进行求解,这里采用顺序交叉算子,对惯性因子w、自我认知因子c1、社会认知因子c2则以w/(w+c1+c2),c1/(w+c1+c2),c2/(w+c1+c2)的概率接受粒子本身、当前最优解、全局最优解交叉的父代之一(即按概率选择其中一个作为父代,不加权)。 算法设计的关键在于如何向表现较好的个体学习,标准粒子群算法引入惯性因子w、自我认知因子c1、社会认知因子c2分别作为自身、当代最优解和历史最优解的权重,指导粒子速度和位置的更新,这在求解函数极值问题时比较容易实现,而在VRP问题上,速度位置的更新则难以直接采用加权的方式进行,一个常见的方法是采用基于遗传算法交叉算子的混合型粒子群算法进行求解,这里采用顺序交叉算子,对惯性因子w、自我认知因子c1、社会认知因子c2则以w/(w+c1+c2),c1/(w+c1+c2),c2/(w+c1+c2)的概率接受粒子本身、当前最优解、全局最优解交叉的父代之一(即按概率选择其中一个作为父代,不加权)。
评论
添加红包

请填写红包祝福语或标题

红包个数最小为10个

红包金额最低5元

当前余额3.43前往充值 >
需支付:10.00
成就一亿技术人!
领取后你会自动成为博主和红包主的粉丝 规则
hope_wisdom
发出的红包
实付
使用余额支付
点击重新获取
扫码支付
钱包余额 0

抵扣说明:

1.余额是钱包充值的虚拟货币,按照1:1的比例进行支付金额的抵扣。
2.余额无法直接购买下载,可以购买VIP、付费专栏及课程。

余额充值