1. No category

Bioinformatik
En fördjupning i
forskningen:
bio- och kemoinformatik
Hantering av stora databaser
Analys av DNA sekvens data
DNA alignments
Bestämming av gener
Genome assembly
Protein strukturbestämning
Analys av proteinexpression
Anders Backlund
Analys av protein-protein interaktioner
Anders Backlund
Avd. f. Farmakognosi
Inst. f. Läkemedelskemi
2015.10.01
Anders Backlund
Jämförande genomanalys
Modellering av evolution
Modellering av populations och systembiologi
Avd. f. Farmakognosi
Inst. f. Läkemedelskemi
Vad är det vi vill ha svar på?
Vilka gener?
Vilka arter?
(läkemedel för en art, eller mot en art?)
Vilken data är vi intesserade av?
Anders Backlund
Anders Backlund
Avd. f. Farmakognosi
Inst. f. Läkemedelskemi
Avd. f. Farmakognosi
Inst. f. Läkemedelskemi
Sekvensanalys av en gen
Sekvensanalys av en gen… eller
ett helt genom!
Databaser
Sekvensbehandling och
genomeassembly
(GenBank, SWISSPROT, EMBL)
Homologisökningar
Genomstruktur
(BLAST)
Genprediktion
-Open Reading Frames (ORF)
-Gen prediktion
-Prediktion av andra genetiska element
(glimmer, GeneMark, ORF finder)
Statistisk analys av DNA eller AA innehåll
Anders Backlund
(Biopython, bioPERL)
Genannotering
Anders Backlund
Avd. f. Farmakognosi
Inst. f. Läkemedelskemi
Avd. f. Farmakognosi
Inst. f. Läkemedelskemi
-Homologisökningar
-Motiv-sökningar
Strese et al. BMC Evolutionary Biology 2014, 14:119
http://www.biomedcentral.com/1471-2148/14/119
Page 3 of 13
Strese et al. BMC Evolutionary Biology 2014, 14:119
http://www.biomedcentral.com/1471-2148/14/119
RESEARCH ARTICLE
Open Access
A recently transferred cluster of bacterial genes in
Trichomonas vaginalis - lateral gene transfer and
the fate of acquired genes
Åke Strese1, Anders Backlund1 and Cecilia Alsmark1,2*
Abstract
Background: Lateral Gene Transfer (LGT) has recently gained recognition as an important contributor to some
eukaryote proteomes, but the mechanisms of acquisition and fixation in eukaryotic genomes are still uncertain. A
previously defined norm for LGTs in microbial eukaryotes states that the majority are genes involved in metabolism,
the LGTs are typically localized one by one, surrounded by vertically inherited genes on the chromosome, and
phylogenetics shows that a broad collection of bacterial lineages have contributed to the transferome.
Results: A unique 34 kbp long fragment with 27 clustered genes (TvLF) of prokaryote origin was identified in the
sequenced genome of the protozoan parasite Trichomonas vaginalis. Using a PCR based approach we confirmed
the presence of the orthologous fragment in four additional T. vaginalis strains. Detailed sequence analyses
unambiguously suggest that TvLF is the result of one single, recent LGT event. The proposed donor is a close
relative to the firmicute bacterium Peptoniphilus harei. High nucleotide sequence similarity between T. vaginalis
strains, as well as to P. harei, and the absence of homologs in other Trichomonas species, suggests that the transfer
event took place after the radiation of the genus Trichomonas. Some genes have undergone pseudogenization and
degradation, indicating that they may not be retained in the future. Functional annotations reveal that genes
involved in informational processes are particularly prone to degradation.
Conclusions: We conclude that, although the majority of eukaryote LGTs are single gene occurrences, they may be
acquired in clusters of several genes that are subsequently cleansed of evolutionarily less advantageous genes.
Keywords: Lateral gene transfer (LGT), Trichomonas, Peptoniphilus, Phylogeny
Anders Backlund
Avd. f. Farmakognosi
Inst. f. Läkemedelskemi
Background
The protozoan parasite Trichomonas vaginalis is a human
pathogen that causes the most common, non-viral, sexually transmitted disease in the world, infecting 248 million
people yearly according to WHO estimates [1]. Men are
often asymptomatic carriers of the parasite, while symptoms in women range from malodorous vaginal discharge,
inflammation and swelling of the urogenital tract to
increased risk for cervical cancer, adverse pregnancy
outcomes and an increased susceptibility to HIV-1 infection
[2-4]. Treatment today is limited to two nitroimidazole
derivatives, tinidazole and metronidazole, although failure
* Correspondence: cecilia.alsmark@fkog.uu.se
Division of Pharmacognosy, Department of Medicinal Chemistry, Uppsala
University, Uppsala, Sweden
2
Department of Virology, Immunobiology and Parasitology, National
Veterinary Institute, Uppsala, Sweden
1
of treatment due to resistance has been reported [5]. A
draft genome sequence of T. vaginalis G3 was accomplished in 2007 [6], revealing an unusually large genome of
more than 160 Mbp, encoding up to 60,000 genes in
addition to numerous and diverse repeated regions.
LGT is the acquisition and fixation in the recipient genome of genetic material from a foreign donor organism
without sexual transfer. It offers a rapid retrieval of new
capabilities such as the ability to utilize new metabolites
[7], degradation of chemicals such as pesticides [8] or the
deployment of drug resistance genes [9]. The bacterial
routes for uptake of foreign DNA are well described by
features such as transformation, conjugation and transduction, or by the activities of “gene transfer agents” such
as transposable elements. The mechanisms for eukaryotic
gene acquisition are less well described [10], although one
of the favored hypothesis suggests that the transfer is
© 2014 Strese et al.; licensee BioMed Central Ltd. This is an Open Access article distributed under the terms of the Creative
Commons Attribution License (http://creativecommons.org/licenses/by/2.0), which permits unrestricted use, distribution, and
reproduction in any medium, provided the original work is properly credited. The Creative Commons Public Domain
Dedication waiver (http://creativecommons.org/publicdomain/zero/1.0/) applies to the data made available in this article,
unless otherwise stated.
Anders Backlund
Avd. f. Farmakognosi
Inst. f. Läkemedelskemi
Figure 1 Gene map overview of the genes of TvLF. Overview of the genes of TvLF, in five strains investigated, of T. vaginalis, and the
corresponding region in P. harei. Genes are numbered in order of appearance so that all orthologs have the same number. All details are listed in
Additional file 1: Table S5. Note that P. harei contains eight genes without homologs in any strain of T. vaginalis (genes abbreviated 2, 4, 6, 10–11,
20, and 22–23) and T. vaginalis possess three genes absent in P. harei (genes denoted with asterisk, abbreviated 3, 5, and 19). Additionally, three
genes (abbreviated 14, 16 and 29) are unique for T. vaginalis G3, Pinna and Moz-4, and are caused by stop codons in these strains. Gene classifications
that are denoted by the different colors are according to Kyoto Encyclopedia of Genes and Genomes pathway (KEGG pathway). Genes without suitable
KEGG-classification are categorized as “other function”. The majority of the primer-pairs used for amplifying and sequencing the genes of TvLF are
visualized along with the primer-pair abbreviation found in Additional file 1: Table S8.
P. harei. In previous studies T. gallinae and T. tenax
have been verified to be the two most closely related
species to T. vaginalis within the class of Trichomonadea
[26]. This indicates that the transfer has occurred after the
divergence of T. vaginalis from the remainder of the genus.
A recent acquisition would be in agreement with the
unusually high nucleotide sequence similarity to orthologs of the putative bacterial donor (Table 2).
The genomic architecture of TvLF
The genes on TvLF encompass a stretch of 27 consecutive
genes of bacterial origin, TVAG_243570-TVAG_243830,
Fylogenetisk analys av sekvensdata
DNA / AA sekvenser
(GenBank, SWISSPROT, EMBL)
Sekvens alignment
(clustal W, X, Muscle, Geneious…)
Tree-building algorithms
(parsimony, maximum likelihood, bayesian inference…)
Tree support
Anders Backlund
Avd. f. Farmakognosi
Inst. f. Läkemedelskemi
(jackknife, bootstrap, Bremer-support…)
Tree interpretation
Anders Backlund
Avd. f. Farmakognosi
Inst. f. Läkemedelskemi
(gain/loss, insert, deletion, traits & characters, trends)
Fotosyntetiserande organismer…
Anders Backlund
Anders Backlund
Avd. f. Farmakognosi
Inst. f. Läkemedelskemi
Avd. f. Farmakognosi
Inst. f. Läkemedelskemi
RuBisCO
L2
L8
L10
Anders Backlund
Anders Backlund
Avd. f. Farmakognosi
Inst. f. Läkemedelskemi
Avd. f. Farmakognosi
Inst. f. Läkemedelskemi
L8S8
138
E. Svangård et al. / Phytochemistry 64 (2003) 135–142
Dockningsanalys
Protein-protein bindning
Protein-DNA bindning
Statistiska modelleringar
CAPRI initiativet
Critical Assessment of PRediction of Interactions
http://www.ebi.ac.uk/msd-srv/capri/
Anders Backlund
Avd. f. Farmakognosi
Inst. f. Läkemedelskemi
Fig. 1. A molecular surface plot of vodo M (1), vodo N (2) and kalata B1, coloured by polarity (red indicating negatively charged residues, blue
indicating positively charged residues, white indicating hydrophobic residues, and yellow indicating hydrophilic residues). The molecular surfaces
were calculated in QUANTA (Accelrys Inc, San Diego). Each peptide is presented in four molecular surface plots, representing rotations of 90!
around the vertical axis, as indicated at the top of the figure. The modelled structures of vodo M and vodo N show amphipathic structures with
hydrophobic amino acids presented on the surface (even in a polar environment).
Strese et al. BMC Evolutionary Biology 2014, 14:119
http://www.biomedcentral.com/1471-2148/14/119
APG & APG II (III)
An ordinal classification
of flowering plants.
The Angiosperm Phylogeny Group
(29 författare)
Annals of the Missouri Botanical Garden
1998, 85, pp. 531-553.
–––––––––––––––––––––––––––––––––––––
An update of the Angiosperm
Phylogeny Group classification
for the orders and families of
flowering plants: APG II
Anders Backlund
Avd. f. Farmakognosi
Inst. f. Läkemedelskemi
The Angiosperm Phylogeny Group
(27 författare)
Botanical Journal of the Linnean Society
2003, 141, pp. 399-436.
Genetik, mutationer
Cancerforskning
Genetiska sjukdommar
Allelisk variation
1000 genome projektet
Kartlägga alla SNPs
Page 3 of 13
Amborellaceae
Nymphaeaceae
Austrobaileyales
Chloranthaceae
Canellales
Piperales
Laurales
Magnoliales
Acorales
Alismatales
Asparagales
Dioscoreales
Liliales
Pandanales
Arecales
Poales
Commelinales
Zingiberales
Ceratophyllales
Ranunculales
Proteales
Gunnerales
Caryophyllales
Santalales
Saxifragales
Crossostomatales
Geraniales
Myrtales
Celastrales
Malphigiales
Oxalidales
Fabales
Rosales
Cucurbitales
Fagales
Brassicales
Malvales
Sapindales
Cornales
Ericales
Garryales
Gentianales
Lamiales
Solanales
Aquifoliales
Asterales
Apiales
Dipsacales
Anders
Backlund
Figure
1 Gene
map overview of the genes of TvLF. Overview of the genes of TvLF, in five strains investigated, of T. vaginalis, and the
corresponding
region in P. harei. Genes are numbered in order of appearance so that all orthologs have the same number. All details are listed in
Avd. f. Farmakognosi
Inst. f. Läkemedelskemi
Additional
file 1: Table S5. Note that P. harei contains eight genes without homologs in any strain of T. vaginalis (genes abbreviated 2, 4, 6, 10–11,
20, and 22–23) and T. vaginalis possess three genes absent in P. harei (genes denoted with asterisk, abbreviated 3, 5, and 19). Additionally, three
genes (abbreviated 14, 16 and 29) are unique for T. vaginalis G3, Pinna and Moz-4, and are caused by stop codons in these strains. Gene classifications
that are denoted by the different colors are according to Kyoto Encyclopedia of Genes and Genomes pathway (KEGG pathway). Genes without suitable
KEGG-classification are categorized as “other function”. The majority of the primer-pairs used for amplifying and sequencing the genes of TvLF are
visualized along with the primer-pair abbreviation found in Additional file 1: Table S8.
P. harei. In previous studies T. gallinae and T. tenax
have been verified to be the two most closely related
species to T. vaginalis within the class of Trichomonadea
[26]. This indicates that the transfer has occurred after the
divergence of T. vaginalis from the remainder of the genus.
A recent acquisition would be in agreement with the
unusually high nucleotide sequence similarity to orthologs of the putative bacterial donor (Table 2).
The genomic architecture of TvLF
The genes on TvLF encompass a stretch of 27 consecutive
genes of bacterial origin, TVAG_243570-TVAG_243830,
Proteinstrukturmodellering
spanning more than 34 kbp of the 52 kbp long contig
DS113827 in the T. vaginalis G3 genome (Figure 1, Table 2
and Additional file 1: Table S4 and Additional file 1:
Table S5). Although absent from the sequenced eukaryote gene-pool, an homologous region was detected in the
Strain
Isolated
Location
firmicute
bacterium
Peptoniphilus
harei
(contig 0004,
G3 (PRA98)1
1973
Beckham,
United Kingdomeller
röntgen
kristallografi
NMR, är
långsamma
och dyra
metoder
positions 22397–56995, HMPREF9286_0330-HMPREF9
Casu2 (SS-22)1
2008
Sardinia, Italy
286_0294, reverse direction). The TvLF stands in contrast
Moz-4 (MPM4)1
1997
Mozambique
to other LGTs detected in parasite genomes that typically
Pinna (SS-28)1
1998
Sardinia, Italy
are singletons embedded among vertically inherited
kan vara
vilseledande
Tor-A (TO-01)1
2010
Turin, Italy
genes [17,27].
2
A comprehensive comparative sequence analysis of the
T1
1993
Taipei, Taiwan
2
TvLF in T. vaginalis G3 and the putative bacterial donor
P9
Prague, Czech Republic
reveals an unusually high degree of nucleotide sequence
1
Strains used in this study to investigate TvLF.
2
similarity (79-98%), compared to that of typical prokaryoteStrains tested positive for the presence of three randomly selected TvLF genes.
Table 1 Identification of T. vaginalis strains used in this
study
Experimentell bestäming
Homologibaserad modellering
Algoritmer för modellering av
proteinveckning
extremt komplexa
Anders Backlund
Anders Backlund
Avd. f. Farmakognosi
Inst. f. Läkemedelskemi
Avd. f. Farmakognosi
Inst. f. Läkemedelskemi
Lipinskis rule of five
ChemGPS-NP dimensioner
En global, 8D, karta över NP kemiska rymd
Experimental and computational approaches
to estimate solubility and permeability in drug
discovery and development settings.
––––––––––––––––––––––––––––––––––––––––––––––
C. A. Lipinski, et al.
2 aromaticitet &konjugation
Advanced Drug Delivery Review
23:3 p.3-25. 1997.
3 lipofilicitet, polaritet & vätebindningskapacitet
1 storlek, form, polariserbarhet
4 flexibilitet & rigiditet
Hur många tänkbara
’läkemedelslika’
substanser finns
det då?
5 electronegativitet, antal kväve, halogener & amider
6 antal ringar, roterbara bindningar, amider & OH
7 antal dubbelbindningar, syre & kväve
Anders Backlund
Anders Backlund
Avd. f. Farmakognosi
Inst. f. Läkemedelskemi
Avd. f. Farmakognosi
Inst. f. Läkemedelskemi
8 aromatiska & alifatiska OH, omättnad, LAI
Larsson, J., Gottfries, J., Muresan, S., och Backlund, A.
ChemGPS-NP: tuned for navigation in biologically relevant chemical space
Journal of Natural Products, 2007, Vol. 70 (5) pp 789-794
…formalisering…
OH
Vad tittar man på?...
O
O
H3CO
H3C
OCH3
OH
OC1=C(C=O)C=C(O)C=C1
CC1=CC(OC)=CC(OC)=C1
OH
OC1=C(C=O)C(C2=C(OC)C=C(OC)C=C2C)=C(O)C=C1
O
OCH3
Anders Backlund
Avd. f. Farmakognosi
Inst. f. Läkemedelskemi
N
N
O
1b
OH
H3CO
CH3
Anders Backlund
SMILES = Simplified Molecular Input Line Entry System
http://www.daylight.com/smiles/
Avd. f. Farmakognosi
Inst. f. Läkemedelskemi
N
CH3
N
Vilka strukturer är mest lika?
...och sen’ då?
DragonX beräknar de 35 molekyldeskriptorerna, dessa är beskrivande
värden för 35 olika aspekter av molekylen
1
2
3
SIMCA gör sedan med hjälp av dessa 35
deskriptorer en prediktion av var på
kartan över kemisk rymd som just denna
molekyl hör hemma.
Anders Backlund
Avd. f. Farmakognosi
Inst. f. Läkemedelskemi
4
Därmed bestäms molekylens position i 8D
(åtta dimensioner), och denna position blir
alltid densamma.
55 - Barettin
Vilka strukturer är mest lika?
Euklidiska avstånd över 8D!
10.9
1
11.9
3
5.8
8.7
8.4
3.9
4.6
5
11.8
5 - Barettin
12.4
11.7
2
4
Klassificera nya cytostatika…
Strukturer och/eller deskriptorer?
Olika sidor av
samma mynt,
precis som
systematik
och ekologi.
ALKYLERARE
PROTEASOM
TYROSINKINAS
För strukturbaserad approach:
+ direkt koppling till biosyntes
TUBULIN-AKTIVA
För deskriptorbaserad approach:
+ färre ad hoc antaganden
+ lätt att skala upp till stora dataset
+ fysikalisk-kemisk verklighet
Anders Backlund
Avd. f. Farmakognosi
Inst. f. Läkemedelskemi
TOPOISOMERAS-I
TOPOISOMERAS-II
ANTIMETABOLITER
c.f. Schuffenhauer et al.,
J. Chem. Inf. Model. 2007. 47: 47-58.
Rosén, J., Rickardson, L., Backlund, A., Gullbo, J., Bohlin, L., Larsson, R., Gottfries, J. (2009)
ChemGPS-NP mapping of chemical compounds for prediction of anticancer mode of action.
QSAR Comb. Sci. 28: 436-446 (2009).
...prediktion mot MOA dataset.
Spåra kemisk syntes...
ALKYLATORS
ANTIMETABOLITES
TUBULIN-ACTIVE
Anders Backlund
Avd. f. Farmakognosi
Inst. f. Läkemedelskemi
Startsubstanser, • + • ger
en serie, • + • den andra.
TOPOISOMERASE-I
INHIBITORS
TOPOISOMERASE-II
INHIBITORS
Anders Backlund
Avd. f. Farmakognosi
Inst. f. Läkemedelskemi
...passar bra med Topo-II inhibitors, mer detaljer kan erhållas från OPLS-DA* analys.
Data from: Synthesis and biological evaluation of phenanthrene derivatives as cytotoxic agents
by Lee, C.-L. et al. in prep for JMC.
*OPLS-DA = orthogonal partial least squares discriminant analysis
Utvärdera okända substanser…
OPLS-DA som visar...
+
Anders Backlund
Avd. f. Farmakognosi
Inst. f. Läkemedelskemi
Anders Backlund
nAT = number of atoms
Se = sum of Sanderson atomic electronegativity
nBT = number of bonds
Sp = sum of atoms polarizability
ARR = aromatic ratio
Ui = unsaturation index
nAB = number of aromatic bonds
nCar = number of aromatic carbons
Avd. f. Farmakognosi
Inst. f. Läkemedelskemi
…deskriptorer som är starkt korrellerade.
Charting biological activity
in chemical property space
using ChemGPS-NP
Anders Backlund, Rosa Buonfiglio, Astrid Henz,
Elisabet Vikeved, Kuei-Hung Lai & Thierry Kogej
[presenterat i]
Anders Backlund
Budapest, 2015.08.23-27
Avd. f. Farmakognosi
Inst. f. Läkemedelskemi
Prediktera substanserna…
Comparing campaigns…
Probing & expanding model!...
Investigating pharmacological similarity by
charting chemical space
Rosa Buonfiglio, Ola Engkvist, Péter Várkonyi, Astrid Henz,
Elisabet Vikeved, Anders Backlund, and Thierry Kogej.
Journal of Chemical Information and Modeling
– Under revision. ID: ci-2015-00375m
ChEMBL collection
Activity cutoff = 10µM
Human target annotation (EGID)
Anders Backlund
Anders Backlund
Avd. f. Farmakognosi
Inst. f. Läkemedelskemi
Div. of Pharmacognosy
514,257 active compounds
divided in
909 protein target sets
Dept. of Medicinal Chemistry
(Enzyme, GPCR, Ion Channel, NHR, Kinase, Transporter)
ChemGPS-NP v/s ECFP_4…
Mapping…
Kärven
ChemGPS-NP
ECFP_4 fingerprint
Stora Salsta
Sågmyra
263
Sundersknallen
Nylunda
Backavattnet
Jobbsbol
Paris
Anders Backlund
Anders Backlund
Div. of Pharmacognosy
Avd. f. Farmakognosi
Inst. f. Läkemedelskemi
Dept. of Medicinal Chemistry
ECFP_4 fingerprint
116
667
7
ChemGPS-NP
Berga
Trollemölla
Flyemyra
Mapping…
First stage of activity-mapping:
–––––––––––––––––––––––––––––––
80 defined activities
21669 compounds retrieved & mapped
………
19508 with unique activities
2161 exhibiting ’polypharmacology’
Anders Backlund
Anders Backlund
Div. of Pharmacognosy
Avd. f. Farmakognosi
Inst. f. Läkemedelskemi
Dept. of Medicinal Chemistry
Anders Backlund
Avd. f. Farmakognosi
Inst. f. Läkemedelskemi
69 compounds with experimentally demonstrated activity against
Leishmania, at concentrations <10µM, obtained from litterature & db’s.
The corresponding chemical property space defined by this set,
represented by 183779 computational nodes (in first 3D).
An example from 63rd GA 2015
anti-Plasmodium,
2262cpd
anti-Leishmania,
69cpd
opiod receptor µ,
648cpd
Anders Backlund
Anders Backlund
Div. of Pharmacognosy
Avd. f. Farmakognosi
Inst. f. Läkemedelskemi
Dept. of Medicinal Chemistry
5-LOX, 1427cpd
12-LOX, 98cpd
15-LOX, 102cpd
An example from 63rd GA 2015
O
An example from 63rd GA 2015
O
OH
OH
Amorfrutin B
OH
OH
CH3
O
CH3
CH3
CH3
Amorfrutin B
CH3
O
CH3
CH3
CH3
”Natural PPARγ agonist with potent
glucose-lowering properties.”
SMILES:
O=C(O)C1=C(O)C(C/C=C(CC/C=C(C)/C)\C)=C(OC)C=C1CCC2=CC=CC=C2
Anders Backlund
Avd. f. Farmakognosi
Inst. f. Läkemedelskemi
ChemGPS-NP position:
0.417125 1.629921 1.938134 0.357786 0.624139 -1.511832 -0.382870 0.014741
Anders Backlund
Avd. f. Farmakognosi
Inst. f. Läkemedelskemi
O
OH
OH
CH3
O
CH3
CH3
CH3
An example from 63rd GA 2015
An example from 63rd GA 2015
O
OH
Amorfrutin B
OH
Amorfrutin B
CH3
693 PPARg-agonists
O
CH3
CH3
CH3
49 closest compounds in ChemGPS-NP chemical property
space are coded for activity on PPARγ or PPARα.
Number 50, 51 and 52 are known as AChE inhibitors:
O
O
N
Anders Backlund
Avd. f. Farmakognosi
Inst. f. Läkemedelskemi
O
O
Avd. f. Farmakognosi
Inst. f. Läkemedelskemi
OH
OH
N
H
OH
Anders Backlund
O
O
N
O
O
CH3
CH3
Anders Backlund
Avd. f. Farmakognosi
Inst. f. Läkemedelskemi
Anders Backlund, Prof.
- chemography & phylogeny
Cecilia Alsmark, Assist.Prof.
- bioinformatics, LGT
Christina Wedén, Dr.
- fungi & fungal compounds
Anna Koptina, Dr.
- biological testing
Muaaz Alajlani, Dr.
- TB & ChemGPS-NP
Kuei-Hung ’Momo’ Lai
- bioactive compounds from fungi
Astrid Henz
- chemography & phylogeny
Åke Strese
- LGT, Trichomonas
Elisabet Vikeved
- LGT, Leishmania
Josefin Rosén, Dr.
- chemography, 2009
Catarina Ekenäs, Dr.
- ethnobotany, 2008
Sonny Larsson, Dr.
- phylogenies, 2007
Petra Lindholm, Dr.
- screening, 2005
…………………………………………………………………………….
Thierry Kogej, Dr.
Rosa Buonfiglio, Dr.
Johan Gottfries, Prof.
- AstraZeneca R&D
- AstraZeneca R&D
- Gothenburg University
O
CH3
Researchgroup for
Molecular Pharmacognosy – 2015.08.26
Support gratefully acknowledged!
Anders Backlund
Avd. f. Farmakognosi
Inst. f. Läkemedelskemi
O
O
O
O
CH3
O
N
O
OH
O
O