6、The Idea 页面
http://www.fight-malaria.org/ind ... le&id=99&Itemid=108
The Idea
我们的想法
To find novel targets by docking known inhibitorsinto structures of malarial proteins 为了寻找新的目标,将已知的抑制剂对接到疟疾道白结构中去。
The target proteins will be processed in order of structural accuracy and reliability: 目标蛋白质的处理按结构的准确性和可靠性依次进行:- X-ray crysal structures
X射线晶体结构 - Tropical Diseases Initiative MODPIPE/MODELLER homology models with active site
热带疾病 MODPIPE/MODELLER 同源模型活性部位 - Tropical Diseases Initiative MODPIPE/MODELLER homology models with score > 1.0
热带疾病 MODPIPE/MODELLER 同源模型,蛋白质评分 > 1.0
Similarly, the ligand small molecules that will be docked into the protein structures will be processed in order of importance:
同样的,将被对接到蛋白质结构中的小分子配体的处理过程,按重要性依次进行: - MMV MalariaBox best candidate hit compounds [200 drugs + 200 probes]
MMV(疟疾药物开发基金组织) MalariaBox 最佳候选命中化合物 [200药物 + 200探针] - Remainder of the MMV/GSK/Novartis/St.Judes hit compound list
剩余的 MMV(疟疾药物开发基金组织)/GSK(葛兰素史克)/Novartis(诺华)/St.Judes(圣犹达) 命中化合物列表 - FDA approved drugs (these have already passed clinical trials in patients, compiled in the NPC)
FDA(美国食品药物管理局) 批准的药物(这些已经在患者身上进行过临床试验,已纳入 NPC 数据库) - ChEMBL Bioassay compounds not included in lists 1-3
ChEMBL 生物活性化合物,但不在 lists 1-3 内 - ZINC Clean Drug Like diversity subset
ZINC 数据库,清洁制剂类,多样性子库 - Head-to-tail cyclical peptides [CycloPs]
首尾相连环肽 [剑水蚤] - ZINC Clean Drug Like full library
ZINC 数据库,清洁制剂类,全数据库
List of planned experiments (and why we need your help!):
计划试验列表(我们需要您的帮助!)
Paper
| Exp
| Target
| recs
| Ligands
| mols
| tautomers
| Vina calcs
| A
| A
| Xray_all [140]
| 140
| Xray_ligs [41]
| 41
| 176
| 24,640
| A
| 1
| Xray_all [140]
| 140
| MMV [400]
| 400
| 852
| 119,280
| A
| Ap
| Xray_all_p [a]
| 138
| Xray_ligs [41]
| 41
| 176
| 24,288
| A
| 1p
| Xray_all_p [a]
| 138
| MMV [400]
| 400
| 852
| 117,576
| A
| 2p
| TDI_site_p [a]
| 28
| MMV [400]
| 400
| 852
| 23,856
| A
| 4p
| TDI_good_p [a]
| 1,425
| MMV [400]
| 400
| 852
| 1,214,100
| B
| 8p
| Xray_all_p [a]
| 138
| NPC [2,646]
| 2,646
| 6,309
| 870,642
| B
| 9p
| TDI_site_p [a]
| 28
| NPC [2,646]
| 2,646
| 6,309
| 176,652
| B
| 10p
| TDI_good_p[a]
| 1,425
| NPC [2,646]
| 2,646
| 6,309
| 8,990,325
| C
| B
| Xray_all [140]
| 140
| MMV_all [18,924]
| 18,924
| 55,943
| 7,832,020
| C
| Bp
| Xray_all_p [a]
| 138
| MMV_all [18,924]
| 18,924
| 55,943
| 7,720,134
| C
| 3p
| TDI_site_p [a]
| 28
| MMV_all [18,924]
| 18,924
| 55,943
| 1,566,404
| C
| 5p
| TDI_good_p [a]
| 1,425
| MMV_all [18,924]
| 18,924
| 55,943
| 79,718,775
| C
| 12p
| Xray_all_p [a]
| 140
| ChEMBLassay[4,262]
| 4,262
| 15,366
| 2,151,240
| C
| 13p
| TDI_site_p [a]
| 28
| ChEMBLassay[4,262]
| 4,262
| 15,366
| 430,248
| C
| 14p
| TDI_good_p [a]
| 1,425
| ChEMBLassay[4,262]
| 4,262
| 15,366
| 21,896,550
|
Abbreviations
缩略语 |
| Xray_lig | All P. fal. proteins that have at least one X-ray crystal structure in RCSB with a co-crystalised ligand (one structure per protein) | Xray_all | All P. fal. proteins that have at least one X-ray crystal structure in RCSB (one structure per protein) | TDI_site | All Tropical Diseases Initiative MODPIPE models with active site (one model per protein) | TDI_good | All Tropical Diseases Initiative MODPIPE models with MPQS score > 1.0 (one model per protein) | DISTILL | DISTILL models of all remaining proteins in the P. fal. proteome (one model per protein) | MMV
| Medicines for Malaria 'Malaria Box' collection
| MMV_all
| Medicines for Malaria full collection of hit compounds from GSK, Novartis and St. Judes Children's Hospital screening studies
| NPC
| NCGC (NIH (National Institutes of Health) Chemical Genomics Center) Pharmaceutical Collection (NPC) of approved drugs (global)
| ChEMBLassay | PubChem/ChEMBL bioassay compounds not part of GSK screen, showing inhibition >90%, in whole P. fal.assays | Xray_lig_wet | Xray_lig with specific crystallographic waters included | TDI_site_flex5x | TDI_site with conformational flexibility (5 conformations each) | TDI_good_flex3x | TDI_good with conformational flexibility (3 conformations each) | ZINC_CDLdiv
| ZINC version 12 Clean Drug Like compounds (subset #13) diversity subset (Tanimoto 90%)
| ZINC_CDL | ZINC version 12 Clean Drug Like compounds (subset #13) full collection | CycloPs_HT
| CycloPs head-to-tail cyclical peptides
剑水蚤首尾相连环肽 |
After completion of experiment 7 we will have achieved our primary goal. Thereafter we will improve the search by allowing taregt protein structure flexibility. Lastly, we will use the BOINC infrastructure to search for novel compounds using available X-ray structures and libraries of small molecules. 当实验7完成后,我们将实现我们的首要目标。此后,我们将提高通过允许目标蛋白质结构搜索的灵活性。最后,我们将使用 BOINC 的基础资源,来搜索新的化合物,使用小分子的X射线结构和库。
FAST and LIGHT approach: Due to the staggering number of calculations that are planned, we soon realised that we would not have the data storage capacity to hold all the results. Instead we'll keep only two small pieces of information from each docking run: - the best pose docking energy
- the Autodock Vina seed number
AutoDock Vina has a convenient function that allows one to start the docking run from a specific seed. By inserting the previous seed used during the virtual screening study, we can recreate the final docked pose, without having to keep huge amounts of data.
How is this different from exisiting research? To our knowledge, only the WISDOM grid project and the GoFightAgainstMalaria BOINC project lead by Alex Perryman (Scripps) are directly aimed at finding novel drugs that could be active against malaria. We had initially planned to do very similar work, but on 16th November 2011 Alex started his WorldCommunityGrid project aimed at finding novel compounds that could be used to combat drug resistant malaria. Details about the targets he is using can be found here. This will involve identifying compounds that may be effective against malaria, and then purchasing and testing these hit compounds in the laboratory. 据我们所知,只有 WISDOM 网格计算项目和 GoFightAgainstMalaria BOINC 项目是由 Alex Perryman (Scripps) 主持的,直接目的就是寻找能够有效对抗疟疾的新药物。我们最初曾打算做非常类似的工作,但是在2011年11月16日,Alex Perryman 启动了他的 WCG 项目,目的在于寻找能用来对付抗药性疟疾的新的化合物。他正使用的目标的细节,请看这里。这将有助于识别能有效对抗疟疾的化合物,然后在实验室内合成和测试这些命中化合物。
While Alex is docking millions of compounds against a small number of target proteins, we are asking if we can find the target protein that is inhibited by a specific compound. In essence it's the same question - backwards. 当 Alex 正在将成千上万种化合物与少数目标蛋白质配对的时候,我们都在问,我们是否能找到一种特殊化合物能够抑制的目标蛋白质。在本质上,这是相同的问题——追溯。
We know from the massively high throughtput assay results from GSK (Tres Cantos), Novartis-GNF and St. Jude's Children's Research hospital that there are 18,924 compounds that inhibit the Plasmodium falciparum parasite in whole cell assays. But we don't know where each of these compounds are affecting the parasite. So in essence we're asking the question "can we find the protein that is inhibited by this compound?" nearly 19,000 times. 我们从 GSK (Tres Cantos)、Novartis-GNF、St. Jude's Children's Research hospital 的大规模数据吞吐量的化验结果中发现,有 18924 种化合物能够抑制恶性疟原虫寄生在细胞的中。但是我们不知道这些化合物如何对寄生虫产生影响。因此,从本质上说,我们正在做的工作,是问“我们能否找到通过这种化合物抑制的蛋白质?”约19000次。
To do that we need structural information about a whole genome... something that is just not available yet. What we do have are 141 very good X-ray crystal structures (PDB) and models for 4,657 plasmodial proteins (of which 818 are similar to known active sites) [Nat. Biotech, 2009]. So we can be fairly confident about 87% of the proteome. Beyond that we'll use other modelling software (DISTILL) to propose structures for the remaining 13% of the 5,363 proteins in the plasmodial proteome. 要完成这项工作,我们需要一套完整的基因结构信息……只是目前尚未问世。现在我们有的,是141种很好的X射线晶体结构(PDB)和4657种疟原虫的蛋白质模型(其中818种和已知的活性部位相似)[Nat. Biotech, 2009]。因此,我们可以相当有信心
We will also use Autodock Vina to do the docking using our own BOINC distributed computation server. In addition, we will use eHiTS on a PS3 cluster to corroborate the findings from the Vina screen.
The result of our work will be a list of proteins that are likely to be the enzymes/receptors targeted by each of the 19,000 hit compounds. These will each have at least one known hit compound. After confirmation in the laboratory, these proteins will then make excellent targets for further drug discovery and development, using the skills of medicinal chemists to optimise the interactions found by the lead compound. Hopefully in the future this will lead to a novel drug that inhibits an as yet unknown protein target in malaria.
|