From 4b7e0d51ae66541f6e48ffe6a62b2b10ea73c8f0 Mon Sep 17 00:00:00 2001 From: Tanya Strydom Date: Tue, 4 Jun 2024 11:26:07 +0100 Subject: [PATCH] =?UTF-8?q?=F0=9F=A7=BC?= MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit --- docs/_tex/index.tex | 12 +++--- docs/_tex/references.bib | 37 +++++++++++++++++ docs/index.docx | Bin 819630 -> 819633 bytes docs/index.html | 20 ++++----- docs/index.pdf | Bin 943595 -> 943630 bytes docs/notebooks/model_qualitative-preview.html | 2 +- docs/notebooks/model_qualitative.out.ipynb | 10 ++--- .../notebooks/model_quantitative-preview.html | 2 +- docs/notebooks/model_quantitative.out.ipynb | 6 +-- index.qmd | 2 +- references.bib | 38 ++++++++++++++++++ 11 files changed, 103 insertions(+), 26 deletions(-) diff --git a/docs/_tex/index.tex b/docs/_tex/index.tex index 240629f..bde9588 100644 --- a/docs/_tex/index.tex +++ b/docs/_tex/index.tex @@ -219,7 +219,7 @@ 1% }% } -\date{2024-05-17} +\date{2024-06-04} \usepackage{setspace} \usepackage[left]{lineno} @@ -330,11 +330,11 @@ will aid them in being able to select the correct model to help them to reach their goal. In order to be able to make informed decisions it is important that one has a strong grasp of exactly what it means to -`code'/define a food web Section~\ref{sec-network-anatomy}, a clear +`code'/define a food web (Section~\ref{sec-network-anatomy}), a clear understanding of why one wants to predict a food web -Section~\ref{sec-network-why}, and ultimately one needs to be able to +(Section~\ref{sec-network-why}), and ultimately one needs to be able to asses and evaluate which model family is going to best match up with the -goal of network prediction Section~\ref{sec-network-build}. Here we +goal of network prediction (Section~\ref{sec-network-build}). Here we specifically aim to look at not look at only the performance of the different models but also initiate a (thus far lacking) discussion around how the interplay between the language used to define networks @@ -451,7 +451,7 @@ \subsection{Putting the parts together; what does it research programmes (or even practical needs) that have been driving the construction of them. -\begin{tcolorbox}[enhanced jigsaw, toprule=.15mm, leftrule=.75mm, breakable, rightrule=.15mm, opacitybacktitle=0.6, arc=.35mm, title=\textcolor{quarto-callout-note-color}{\faInfo}\hspace{0.5em}{Box 1 - Mechanisms that determine feeding links}, colback=white, titlerule=0mm, opacityback=0, colframe=quarto-callout-note-color-frame, left=2mm, bottomtitle=1mm, coltitle=black, toptitle=1mm, bottomrule=.15mm, colbacktitle=quarto-callout-note-color!10!white] +\begin{tcolorbox}[enhanced jigsaw, bottomrule=.15mm, toptitle=1mm, colback=white, titlerule=0mm, colbacktitle=quarto-callout-note-color!10!white, breakable, arc=.35mm, leftrule=.75mm, bottomtitle=1mm, opacityback=0, colframe=quarto-callout-note-color-frame, title=\textcolor{quarto-callout-note-color}{\faInfo}\hspace{0.5em}{Box 1 - Mechanisms that determine feeding links}, left=2mm, rightrule=.15mm, toprule=.15mm, opacitybacktitle=0.6, coltitle=black] There are many ideas as to what are the underlying mechanisms that determine the links between species. The way one chooses to encode a @@ -726,7 +726,7 @@ \subsection{Model families}\label{model-families} \end{figure}% -\begin{tcolorbox}[enhanced jigsaw, toprule=.15mm, leftrule=.75mm, breakable, rightrule=.15mm, opacitybacktitle=0.6, arc=.35mm, title=\textcolor{quarto-callout-note-color}{\faInfo}\hspace{0.5em}{Box 2 - Assessing model outputs}, colback=white, titlerule=0mm, opacityback=0, colframe=quarto-callout-note-color-frame, left=2mm, bottomtitle=1mm, coltitle=black, toptitle=1mm, bottomrule=.15mm, colbacktitle=quarto-callout-note-color!10!white] +\begin{tcolorbox}[enhanced jigsaw, bottomrule=.15mm, toptitle=1mm, colback=white, titlerule=0mm, colbacktitle=quarto-callout-note-color!10!white, breakable, arc=.35mm, leftrule=.75mm, bottomtitle=1mm, opacityback=0, colframe=quarto-callout-note-color-frame, title=\textcolor{quarto-callout-note-color}{\faInfo}\hspace{0.5em}{Box 2 - Assessing model outputs}, left=2mm, rightrule=.15mm, toprule=.15mm, opacitybacktitle=0.6, coltitle=black] Although understanding the underlying philosophy of the different model families is beneficial it is also important to understand in what diff --git a/docs/_tex/references.bib b/docs/_tex/references.bib index 4c8d247..69eae42 100644 --- a/docs/_tex/references.bib +++ b/docs/_tex/references.bib @@ -1,3 +1,19 @@ +@article{adhuryaNovelMethodPredicting, + title = {A Novel Method for Predicting Ecological Interactions with an Unsupervised Machine Learning Algorithm}, + author = {Adhurya, Sagar and Park, Young-Seuk}, + journal = {Methods in Ecology and Evolution}, + volume = {n/a}, + number = {n/a}, + issn = {2041-210X}, + doi = {10.1111/2041-210X.14358}, + urldate = {2024-06-03}, + abstract = {This gap in knowledge regarding ecological interactions has prompted the development of various predictive approaches. Traditionally, ecological interactions have been inferred using traits. However, the lack of trait information for numerous organisms necessitates using phylogenetic data and statistical insights from interaction matrices for prediction. Previous studies have overlooked the validation of model-predicted interactions. This study used a novel method in predicting ecological interactions using a self-organizing map (SOM), an unsupervised machine learning algorithm. The SOM learns from the input interaction matrix by grouping the nodes into output layers based on their interactions. Subsequently, the trained model predicts the interactions as scores. To distinguish between interactions and non-interactions, we employed F1 score maximization, setting scores above a specific threshold as interactions and the remainder as non-interactions. We applied this method to three unipartite metawebs and one bipartite metaweb and subsequently validated the predicted interactions using two innovative approaches: taxonomic and interaction recovery validation. Our method exhibited outstanding predictive performance, particularly for large networks. Various binary classification performance indicators, including F1 score (0.84--0.97) and accuracy (0.97--0.99), indicated high performance. Moreover, the method generated minimal predicted interactions, signifying low noise in the predictions, particularly for large networks. Taxonomic validation excels in metawebs with a connectance {$>$}0.1 but performs poorly in metawebs with very low connectance. In contrast, interaction recovery was most effective in larger metawebs. Our proposed method excels at making highly accurate predictions of ecological interactions with minimal noise, solely utilizing input interaction data without relying on traits or phylogenetic information regarding interacting nodes. These predictions are particularly precise for large networks, underscoring their potential to address knowledge gaps in emerging extensive metawebs. Notably, taxonomic validation and interaction recovery methods are sensitive to connectance and network size, respectively, suggesting prospects for developing robust interaction validation methods.}, + copyright = {{\copyright} 2024 The Author(s). Methods in Ecology and Evolution published by John Wiley \& Sons Ltd on behalf of British Ecological Society.}, + langid = {english}, + keywords = {ecological interaction,ecological network,Eltonian shortfall,interaction prediction,interaction validation,metaweb,network prediction,self-organizing map (SOM)}, + file = {/Users/tanyastrydom/Zotero/storage/6MDYZGGG/Adhurya and Park - A novel method for predicting ecological interacti.pdf;/Users/tanyastrydom/Zotero/storage/EGHM9LFC/2041-210X.html} +} + @article{allesinaFoodWebModels2009, title = {Food Web Models: A Plea for Groups}, shorttitle = {Food Web Models}, @@ -486,6 +502,27 @@ @article{cohenStochasticTheoryCommunity1985 file = {/Users/tanyastrydom/Zotero/storage/ZGDQ9C63/Cohen et al. - 1997 - A stochastic theory of community food webs I. Mode.pdf} } +@article{cooperDeepDiveEstimatingGlobal2024, + title = {{{DeepDive}}: Estimating Global Biodiversity Patterns through Time Using Deep Learning}, + shorttitle = {{{DeepDive}}}, + author = {Cooper, Rebecca B. and {Flannery-Sutherland}, Joseph T. and Silvestro, Daniele}, + year = {2024}, + month = may, + journal = {Nature Communications}, + volume = {15}, + number = {1}, + pages = {4199}, + publisher = {Nature Publishing Group}, + issn = {2041-1723}, + doi = {10.1038/s41467-024-48434-7}, + urldate = {2024-06-04}, + abstract = {Understanding how biodiversity has changed through time is a central goal of evolutionary biology. However, estimates of past biodiversity are challenged by the inherent incompleteness of the fossil record, even when state-of-the-art statistical methods are applied to adjust estimates while correcting for sampling biases. Here we develop an approach based on stochastic simulations of biodiversity and a deep learning model to infer richness at global or regional scales through time while incorporating spatial, temporal and taxonomic sampling variation. Our method outperforms alternative approaches across simulated datasets, especially at large spatial scales, providing robust palaeodiversity estimates under a wide range of preservation scenarios. We apply our method on two empirical datasets of different taxonomic and temporal scope: the Permian-Triassic record of marine animals and the Cenozoic evolution of proboscideans. Our estimates provide a revised quantitative assessment of two mass extinctions in the marine record and reveal rapid diversification of proboscideans following their expansion out of Africa and a\,{$>$}70\% diversity drop in the Pleistocene.}, + copyright = {2024 The Author(s)}, + langid = {english}, + keywords = {Biodiversity,Machine learning,Palaeontology,Taxonomy}, + file = {/Users/tanyastrydom/Zotero/storage/TPYM2GVT/Cooper et al. - 2024 - DeepDive estimating global biodiversity patterns .pdf} +} + @article{daleGraphsSpatialGraphs2010, title = {From {{Graphs}} to {{Spatial Graphs}}}, author = {Dale, M.R.T. and Fortin, M.-J.}, diff --git a/docs/index.docx b/docs/index.docx index e48ca1abf4bc01271fd9c78400d47c5c50d44cc7..7ca5a1a819241cc96906152ffec6c04484450dc4 100644 GIT binary patch delta 22521 zcmV(?K-a&ngEFy$G7V5m0|XQR0tg5I@lwRG4b%Yx@lwRIx^|Q=K)dA*OW=Qw;&h=e;iEWJ4)bJ|R({ZdFPFjwT}EnA|vyDAy+tmnza+~nTR-zOr^{dA^ijf4Xed8MF8J2 zCzdOo%(4)srS}alOcxC13Kru;<2X-AVgSVl00wS$^CJl|8NxUw`F_&p3pAjK=6B|aWiIctc2 z2LAZ{|2{-}<*<4%<%3A_kKg~Fh6Z2Wi-4s~X^p@3I{iaW3Wr0D=5H3#?g4-F!GPQ# zAY^&XpYlOxc*y7vxi_-Jh?7m6f{GAm0>-a_Zh?#y%nW4-3IyP6p^P0C!t^7sf`C&j zIfFF-jtIv^3nRMJ%S0FyCj+>84VG|!qPd0lQX%_*7lw!1-n}K}LwA$P>8B@SlPfqElDpcuzOk<`xsmnE^vh%5c2RBBwoj;1crsjAxLrI zfuOSTpecHp96~J}2j-#mkLDbINr#D#8EV2#kg!>L&@-BnAG!P{@dr1vXA}tiJMxJC; zXi!?6c_Op81|{$TLJ+j$C0cmREwtNldrchzTyHYSJTOrgO9rM$y4WHVK!(aVmbm^} zEkP}w3`g80_{p-Y<$Gd(J_o@tOJI-4ya8e_EKBnX5_}*Uc%ZB;viaB!K6l08e8OLQ zx?t>h7~m$>c!KkyjqkviB*uJ4`KLe zLP;G~JJtHjQiA5dko(ONxpb6R#%0Q~`wkW{#gihR?^HwbhZoj0+)XAxKPm1b`Uh@P zDPB_fTyt0kTMgQBtU1S0@$i~!Hn~FHpHB(yA(lex+GvdfZRi2Z6mBM9-G0ls(o7cl zJN+6YBSRernY{Xc+!%2*W~!jOg@sC%^h=yG39L8~xCYj4v9g{tL9L(SrFz3iO5Hk} zr@8i^@HoDKAMzW9ZZG<&c67Bu2eVuQb~)!5tx(2L_Q5)#e-GE)DDfmLK?~4PGh9rW ztR0oXp1^TU_}NO3KrMVs%?YScleD>#8EBcFHi}_JAy{dDj+o&Agf-XB(TMChL7JBK z*dmt?rBB@HEZ35(!1y(g$o)|EI2%6y`zsH^Di7jL$^$>2pEnjv<+Ja<|98VWiA#Y& z2I4bQ+Rds!IBHoc=8LqfWO*EkCNpxLMQUXXYkg(T|2kDC=ME2FLY_gqk?->m?QZx! zkxHHUndT&aIcu>FALAFDhtksyNwBJ`QJ;$@9HPIVb2G}*3$%vG8OeQevslfvP}mE- z!e}MNw_{Wqm~}}s_#vl-vWu4T!s0hVhM~5UqHid-I_y3amx&~Q?uk0Z(UIE>cBQi0 zBG5XSh|G@Kae(aOerh%O zE`<`EOT!Ov=>_^U=Cm;brh!>jG~6^bJlXg;SaUoHMf4dc9MqEL0xh3 z#s1HK|KV;PzkE1fni}ebn)El+f7tNVLL12|hRPW+ZVG`sWF6q14Xr`%L2FR8-*ML4 zcmyN&BVNp5tQqnMtz?xn;x!(sxioCmgL4lDFhxgUtYb2Gvhlob4=x?NqU3UJXifX~ zU(<8vM|AD3cr|Fm1o5s=%Fm98Hrm&u;BU8oK5p9)xlT;(SYp=BNxA^>IRjEQzM|8T zrwF*U%!**JX3BBM(E3SN=zPXcLl9;eo8RK=xx3ATUG-rJ*G3S5Vt0`{l|jc_kVZt~ zL+Z(`*Rs`G8XCpz1lmGN?Tae8pjyFQHF;~v1Osl2(Qa2GQZF=r4ApvJHAfMT=QFK; z*}-?ez~?E@Dcl(IBUYFnJM4D5#T7px#jT!`PxqsQue*3z$ja~N%a=0DmQm^Z$_J7B z@deOB{_ek)-qFCi#Z&+3+Yf=mCTVFbY>h5_JLH+T;ekD+H~{!Fy@0Xy`ZRPAKGxJ=qU7uNHp9o>O{@@JlaW zq30`0!~;@x|N0krg=xR~MRX0%#09}O{`>1k4jv%7Dn0^VJIQnM zYk&7Fc^eqIW6wcLDKvvizNAyQMf{1aZ%iT9|wOU=0K*Pkv?}mu!x6`wqP#*m%^+E^j*P!ozpkxQiqPzoTr4Kht0RBuHaX#J-$DKtu(9b=VBn;>}mi8>}?9J=}AJicJpLS{++ z6d)6IN_jck{OPnXR3udXvYg;>Q3M))3qLv|HIpp)d<=R!ibWzn1u5sD_{qBAkJIr+0>(a=%D^uoU{GVqUoDmNFGHrR^!>LA-o_`RV6>cmD>* z4Y=EBIjstf=aN6!my$+mhe&xZO#jY#X;Kaop=~CcG+P~UE6gVfqr!q9Z-OzkBWM9; zhgoTIvX_>*l+tE5wxe@F_4RQgJutb@a>i=KbwN53-h$#8F7@tXTGJoD|HtW8tInDu zs6!RTh_PjS{U@=O%`W?;-!1Hamu>F0FWE6af<^~p?H%BUb}2>DJV1*TEB6m{g5Nv4 z{uO1RRWni)=S!snx}Z>K*RJ+@)9<3R$f(XcDp`pe5)jcRz^km$m1(4r%Kq&xdxp>W{NV+V4N6os4ChI z*urf-^s>8aQ-OPs_neG2=MqN$kibl*8ic4;{ggDaJh-kd${B`vfO5$Kt{u#5S2prS z4l)+4mtyu631=esK8wi`(g`I-z-XfY28{Hxm3T|Eu!{xyLBSEQ_QEIuhLuzlPf1FpSn}^r zw?}@qhAak!$`8YTIt%34il1o~a>mY(f!tN??;oP}pSxIJ>FB(q!tjO1egC!e{t}y} z!n`$U{kY0Jp))7ca#VhM*SI=8J zSoDQO*>lV%{%F^p(o|_XN8!xzb%qiG5r!~-see*8jPX)=5i_zDrq`2;~yu$ z&}F+Km5K6wgDljU&ml{Ik(lMu3YuLGa*-AE83>*dPex`zd0<8T7wkr)nu4Th|vA6*{6O*O5|CNgp+ZUTzaoQz+lmt;FTTdA!=nQ0mh_p8&5xTms zV4E=$NyX2pVKk0?PN1y;K7GE}fDDcr`%vR`j5`PK)|_GAGK3zZsUf@eaVA(u_hy!9E2&d{1^ zMyKw7IlmAWSCmbhq9%aaEIjY)Db>oR;=vMQ3b@EAV4tI6yMqJuYBHEA2IaYPnG*hr zj~lOlX*RP^$&DhTHGERnDGU3y|K7&!(A$C~14PKkt+K@D&aVld{d>o;mTgnwy?0P; z>49ou)9!{}dk4YC90X5T-golV;xV7EK}@lKNP)Vc%kUTi7hoL|Y+z%PJg6j<(1VI} z-E~f;A!^k#;DeZsTdnTgl@bBwkabkTPf^A4QIj?0c%+8X0h6j5(fxi@Th_QyaD^dVX6k0)PCBSk&FgCf;Hu`7@W2rQN3C2hiFFcrOgZ2J@dXMsb(1}x+Bo&hSQVyKyc!~7=4D`$1L0C)& z3QhEPh$ut41h#0>piwQUtOd;fpyGK1q#&)ks8?rBoDo!y zaaj|L$E+d_bRDHKg_LCVo*#)xTD29sj7``RmQ-{3!zzNCTDaxAdvbMq`HgJt#7kyoO zc$a9B)&p;9Qn3t)oO@V%l<-2Qx9+MA@wZYX8e$!k3EDD zV9_kmegJ|8z@KlYsurcBus=u)BLoTH%qv&nBYQ7P1wzM4__iB#V_;g4eZnE@Iq((7_3 zC7oL1=*Ey6MGv)uASL`zu=Rh-b|rHyITm3HwS*m4oI5Y%jpKq2{;v3cJgsE=efRaV*5-BB%9-o)rUr4A8DEhGfwkbT z$1*+42s#Db)uaJtsQyKurTQlK_cP})@`R;`4DzAThYiSHdl-74g;Oi|d6DOg=b2Pc zI3Ym^?@QwoDXGgnZRWjyP$uD?EDlz-nD^Rwjd=!!o!&u+Af}kt#E}Za6GLtrU{kz> z`PW!l*^(G&IS5}eLqnDz92Z_kSQ~H*LF6p49QXtsB!Z! z!I6^$@1p)?Gf2$d(7iDxij}0};pFx3*+p?DTC&m;D$jk%!|)x;7pweKJ7-kz6Mt5@NAjUMavuzVs~ALL_4!Ga8hm_1 zFN|#SacGr^yMx<*BdX4L>lGLI%`S@DAB-SlYZP&JtF?}_%xnfK(09)!w(4{QD zx1)c)T)ONIdjngNYG2n|xIAuS;qt}bfvu^F6x*IC-Jj&*GA?4S7hqi`sk~4NFeNX{ zlP_SNHy07;TctYZNo(tzM~_wKyo08mW|i}WW-Xe-)O+rK-BIb_1GC(6!0%ZhdzhJZ zpy_#^d2X57Zu_(x-xL7?X_bmXjkBl>bF$ny&7EFvjV8YZYu6O4Zq%AC5r5XYQ4MxV z!e<0Ja!ndD>Ouw!HFs~py;!UWupTwn8Po7y59cdhxEUBHS|>1`kd!YM+(< zR@d8T!jHMc7^p^vnu`M{ZV1jMNb)lN^_T9u0!xLn;P-xLkoS|#9};tlt(q4?Tw>$PoqUZ`? zVH=~C#?sl}^DMEnnBjg+QAvHO zmTFMd0y_3Y5Fk(o8C1h#Amp5kbtgVxtSerMJyxur8F}DB1USjHYeW4r!eF-L6owp} zlcEiQ&2J(bJm8G7daq3(7N-2+9S^mCFTb3|b3avb2fM%D&93t9CiUHOQ(QEn-+pR| zOVcEOdszTtz7?A7%pgOS-D0L|!omQa-}}kS=kgs(7x;qSrGJHcE$Msu1L|lKGXN(An#?w>lwuD z$d!r79Td1S-Zt`MW8ACamdP8DX?ORogctnOFC|G(f&;zMc`$$3nh-nPNQf=I>UTH3 zsvs1)+Yt(}5k>>(gji6yHXtQ`E2e{3u|d<#Gmjppw(>#vZpYeR<#4FhT}7uwRb0oZ zKp+%91G7Rm*--ipI(=7XA5$~|{r%PAu(-fWE*?O&fK&->%Bppk3PU3+Bqy;%O$_tR zs82Swnq#{7qC{;)hj|T%_7FhnuZrK6Lzv%t(@#5;%~Glc!Y(kd+9sucTd%hE27aj_ z^RdLysx>DT?Yx+VY0;$-gb6dd`U~tbj}&)e=;t(tG0a1oO%44pHS->o)raD1Rum9z zP2CFbmf#i+6oGij)sXum{tN0%i&u)CO5+%QJ5&XHU>^Af%1Ic42+}QJ}sTLGP+YL zQ7e|y68aIXCo-zgpq<8fBd;%|W}I%oZkl?O(&1hwGY4&>|NMR81{=sx5X&%`z2)DGNX6^|h_wO{+yq zisv!Obkl{E?23sbrR(?~ zM=jA6?oILa3bMXo2)o>Kx=dXo8eoYRfe2kQ5u#qxfmOwj154KWI+e}PhT>65Z8osm z8rgKZb<&G}hh?E1nK$mxZ1A$fj+MK$on*Sr(J6v$6exVH;M!_$^!TeY?}r2W*Rm=l z-LMtQqLr4si3h%uF@=SJ7nqKIci@+$=7&^^1p1IO6Nto*PLopm>{H!W+p15>^y*RP z&M&$XIWZiK0--^*sxT3S-+x~hxP*t)z0DvT80TDnaXB_S2QI!a{lF7WE@=6Oq#MM2qsc5kf9U3I6EAs+Dkka*JaE(Y@hd7 zD^&(TGaGu?Vp;T(D%6Alj}$8R^x^|N?lVMxW7eUjiMb38S8i&T%Ur!)CN%Nj_EY!t zOm16$sOEI);?(81f+6$aL`Cbo46NvbTB(V$hp#x3CTXJTf-6Z(hBTPuMV!U0V7^(Y zd1!ve#thnnkoSs20v+oVRB=iq)*>2djPoSC!kRgvV560_r4(@k?QlElB~l5-X0|jW zDcxBV*G+Uw62Ry0-wg#eull&^<3 z+}X+E4g%La3|qf)B6QKGQ9!insy*yhf2;PYdmN5=FMH^)3fdg|F88u@@`FGzPK~F3 zxedvmx<9I-eb6SVe@C}{LDXKh=1M$VH`z%w!v-my${WBufs+jkx2d9=U;77PnH>hI zO8lpUv6Y_!skh8ys@ych6m(R2OYFnaz@4}J(PDS)tM)&Vi1u#q6v&K_`7e>JN z48Ad82rN*je=gp==0f=bSW`j#w{Q%9meigN&$rs>-MPhxH?)1tbOAX`H31iZkkma7X=w6P+q&N-j@LqG}Z$C!#b{9CRxvidd z73+FJDd4;D;V3=y)UhZ;Ebkt%y`9iWxn2GCMAXxwdmD){quuC0zmMR9f!Q>{yARQ_WVV#Bh%yyZ?|se_6p=8;x_N@RFi^lrH2iY-lIj}!f?xx4?F zktV1Y1?juoEw?~dG0hQBnq!nD*9+IAOLUz6_#m(OS0E?bEs&sp>a!%4debX)HD;B+ zcivK^X3-s6n?s(MBh+@@CfhayRhWlW7cNVSvR~ikNmRC>1l1l<)kLZD2vzM!O~R!t z!;)Tg;R%_8r~nOTvzhbrM_N2s!%)i4dGUWM#3CGdMwUj9eeM3JJFYEPTTq#*McAVf zh+N0Cg-_p?m>kf5s)pl%HVORJ+mqLRC5dWz?H5Je9Z^J|n3pyQ}Kdn2DAro^gj zNp+HW?nRz3L4ifC8_d;T7Mpsd{j62(AngCIm5QZNhpfYYs{zd{;8BpsX0p|KUuB6f z@h8N(URCk!`4-LqROS#t?hLCZp=m92i+KRbPPU9q_IwrBnH5g0OP$C}koStfm`h!Y zxB3<7UNb~PAb)~UrFP~J8AVRx!_;&n5@L4A+o0d8xh_15<99Hpd{l`3hSDzRC-7k5+ zju@e%sEx<4Yx`V0Mhlp~l)HB*)NV>+VQUoCcE~ev!-KrZNNL7DBqG3*1gM*YRGSQ)RVMyxc`N(wU&!E;qC!n z6pXM~CQ{DzGwo4-a%1`!rnDn_vZbSC^GITr$uW%eOD(}# zH(*a7>~8%dwNF0^PNX|n9K&|Kzg44?!(r;gmpm&q?}U;&{M5E6{V`1XxY*UUgwmD{ z^&7W3LT;59tl?wWulF|koFj6;eL9tN1&`~*Hl1_dj?gj{Avx?uJjB1<(k(eIX1dXT zeC`;AtTHbU*FqWZ<}n@Drr2x!BmLENx3$%1H|Mr#OxH(23>Ds;E0F=>+ouaIk}@2a zsXRE2^LbPZ@R*F>Z))TbsgZfA-!W`h$uYZ)ALXsXQi?)2Cl$kHTRWlF&2Sf#QA>w= zMEcm#8M(h@*li(gNKrT@9z_YE4qtkIJiskHhRv&RDr4W~kw%?`O7${~uoa4Wa+{}p z40HU(tC|{WED6qrH0mScTDs&{=z`#581S$C$?=(x8ggUFQh#y843R+Y60N(j94!OO zPG(qCg^ch*qYlE479u{11MrP2e$(KN&yO}bJ|3u7*YQIv++3leq)=*$2Q&G@$pT5gBT`=#T+|K_&4N!Ct4|DEdU(Z_#!M~Uj&IER8?*ivL&?{Gz)Y0St~vFN zJce!ig?q|RSZuoaQm@D|{uVlaC}o6zIt10?F?Ag0=No7G#+i;!jcg2$`ef#;W#*{g zGB_CD?}>o6l(YKsO!|J7e4(_Zr2}?imcq{|RxNVyk+Su!sE-%A0kXX1asDGhC&b?C z2=fSng>UHya=dGD3?n}x_RN%$zR(zC|LIxuKIIYr=z{Jot^NjPPO-Uv%huoR9>YEQ zNIF|)zxoE-&>(EXq=0QG;RD_GAnd03AYm&|nBZmnT#zbFF9}>Hb^IIHqJDos>*cR; zT-|AJtP!{HGQ!AxtuOXyi9`A~p?_2QS6>=PoG_s&eD@5#5^2cs1M2wcspJ1f-CEOjw`v5$;X(HW#p}8G6NEqdNDRVOsIXPU=HkE#>WScyD zOp)fme@3kRsd$Ym>^Q~y?INyn$=8-+m50Be3MB=zF(1=^7pMy51PGgnqXzQ(7ZBWl zUg`Crd!qvw8TfcOc|8O|&F^IPB7Z{Vxi5JbzGL}fm7i+otVL;X!%Bmu>k%5Vmsg3$~}?~)scg>lU~K4NUP6Js?^})8zaoMP{*NFChiVy zkElB1tyf%s2Qs~9rgxR8rawM>I>60 z>I)Zt2iBr4QWmY@Bq!@YE-x-)pdsI2*8Y|Om1@TcO5t?~Y0bADHEq1Af1M9;;ty%(7JjsCQmlrm_o**S2OY zy2IIAmc7_+I}t;x&5AX0FLm`0%UC9XJwzHq3nz{7!h!u zB;>MxydPQQ5C}Xfhs<<$q3$sW9#YB%@)5&zOH)^zA`{699g}5`qM0!bj+9n{Lnfp> zS&THtMy^09Z_E#y_%e7Tk72;nd={qxtHtIFv2tjRdbe!%H0oco*~1gYzu6IO8``hp z{+XHQ*=n(_8K6@)A$9B6ODbtPU3D7g)zh(3j>Jp&&c}iPFJ$;w4&PhCXUSnHrF+=XS37sB) z7me%o^mpIv-%X7vR0yn@e=hnKEm2hZ3!pp>CWmR7R#ikwl0R=}wpF$}wuxzu z%}W#t1Hd5T34MhvVv1I$rmMn?+a9-bam`J?octu3Fq!F`vw4DOqm2G!wgQfSqtBC~ z+fUxe>{b^%tH&6AQk{(kvX?OMB;&|N1lCHLabUX}{YAv*%zQy{NTY}F{Ba6KTRcO1 z>ZsGtMqk#ZsAJGQj5^*>cgQkGJP5>v1jUI_)EH$$MWDTHUoy<-=k}M zi#4@d2K_dF`o2u8%r1V(mo;dsFKf{I7B2f1F54Y0>pyp1>w9lW5kuK$PQiQ_t1)xpU<@XU^5Iz>>b@a!En+%S{bXLISL!o{5LHU)nFc-KLWF zbi+c_o37mF&ih>5LYiO+>VzUv9p|c7{!%U@agE?7v>N#Jp!_w9LDwUHo(epg1N*~# zA$ut<{?)L2c_8q3Y`OT3VaLI+fw1HKY<1$$^wlla!2WqeUHS0Z(~j}D(% zr@$C_<*IE0>uPh`zOIOWw`)YLcLUA86FE!Kn1S^H+o0%rh2RBrSc=`z`#^g%Gi^nU z$pT+cxhC)_Y%AQ!`;0EpDQj&Csq;CpS#=iaR$X`>eJ!Twf*F%LI`R@umfe75hcloj zvuhA5sLwvW=j%h>V2m;us9QNKL|xH{y^pp{<&zvd4U|J?a7h?+9G@wEmUnC{ZRXQED!TU=E|o1s!{uWRbz(0s z5(ONHRQaugIX$3ZkULE`7**%ad#=4?=o-k3JD3^;Z*;pu zQ{9GrbJd}2x!1IRGY6grFu>W0dkyW)W(eQ`Bm{bs%9_%JdI#x3+E2queVDKjyYVfb zyhXPVwr-}B_W&tzR&~*sHFvI6`KRd6p>PTwW(5~H3$8>SDAZ~8&O5Mq3b!Be=9H_n z;;e@6O_a;B3Vjebdcd%NQ$J_h?FtnYr63+ty zOAPr$3vM*zT&U&lQ0rA*VoPc&F*dEoahq|z!Z@WFxg;Ek-p0Z^OC+{f$C84)>HtXOZ<|^BW*^ zbmNz zL1M()LC^UbomLeHt=n6hr*-QPtEYvx8qNk9 z-?00CX9>p(@7(dccL22woR|4?9P^wn5_yy@4R;Q&!#oyXh6i814eW<*x0{c~{OXRU zKv@w?N__5h*qNMEPMboVpR{jKz2yZcJQznKy(bLMb%HtxIDJT**cxMC7c-GK9(pSx z6on5UyFr-IumgM!Yk@0J zr1X^=AGLx_^9a*qTEU%lA`E&%O!f~$+KHI|uWt<04#MMdp@&$&KJYFp0mmJ)0%lh*5pV5TLc0p>%bq*-Q+ z((+)r$z~(jhj;S0elC0tmMcb)@DYw$!m&ZPmymB&eICL-2eq@$U$pCuI&Q;?sDg@`D*R4&5 zBDDhPQYTC5Z0~EJtPs1UJ_WrJ=bK{CJ*nmCZDJSx~~(e9)}g92GS5G zj=79OZf$H#4#VS?tF9@#k&#@&M#3*-m}&PkW~4GWh@^n)BoJk?KXt5&ywuo#32fu4 z^;|%$7Ywa2uV}Y3=pEvt=oGnstXSaTU|uTntT==Vt1=wZb%ml#HQ;qLv#QG}(Cjl4 zYC$Wx!py%Me?Cn(cg}2_gi_|#_v_^)Z!sGX6wuUvI7P@E*$Oo} zVF#VV!<6WYh!hr^Fn)p_&;R*9e8k(mmWhO8wh(bF)oIm94YW*4`u!4 z{75p1p~E;FJ%NA*9Jd89HJ(X0K3(J zKA#fDaz1LUb6eCwa5U(DClxvBB>}$cSRNXrY+X*1k1(WS3*2>Dv*Jd-^f=048ru0m zKtud}Z;RYQCzyv*+(6~uy&s~C`c!Z%L zgS>#}-X_y73uAOo3268?u&23>1uoFPMl(2B$e2^P$SIm27}vWUk>O+{8^#0^EUqZI zsc2;-?KM+Waf0rDKSc$Zr7)yEi??Hu`14p`<)jZrGipy*k!r@?6bEW9o>o%IPiSSZ zT>rFEW@ccUg~DmhIB*v<<>9qmFLazb4>y2zvfqAFlSh@FOl)GP{Dx6oi_o*xm322M zDgM`No}u@G1^;B2$y>9^IIiDCn=>gbQWUTQ>#6pW?8d%-X30Fmq%FrJM>Lz4J7{Vr zO?qXmO5Obom>oLb{cZ+2?E(&HD0H=F#q)we$UxpaFNOUInd4zrL%DQ!2{Mfmuu)PW zwS%Mq*})PKuD9NS{B$a%YIT#U$AjAj`c~ENzAJZpnqhZauYm_h&dI$b=PSyAB6e&y z$Yx2%?PYp@I=pzx(ObH_@F~~J>ZbRa;bE4P&9=yyo6QgYC)kf^DQC3Y-fweFv?h-6 z1Wd7Te3AHB2s>1>Xkis1FV6#{Q?)t=k&9w17&tkoldI`*$lFv^9dCkwT z>ImMQTE;3=C3%uZiZ9W=CL$rnKT{=H>^0>F`*kpE1?zf=V#cAICtM)r_~iAMj$uqT zd5bjEa_X1mQ~B{F00CWNVGEWvF0CA*;&MDh$S`zzk`Bn1A`#V4n+F$yEojxM-0`ax zlq}|dVK%j&6<*&o)fw^HMCEYALSdH=qb2l{jg)Q7@ zg{uRNa`dpZ!qb7WP#}frj1>(?kXizsOYMk%qBEJt5{iUapjt<524)`0JKXNQc*n1U ztu6At`Zx}C^8!cFacspz&?(?Xm@euwVqjBuG1hh~2pY!eiGz?)nxkh8&P5pjpz_VK z$T#(2Skz?<8qnM3@Fo0~I{0qw=v=34Z7i8z`*pQQG$)plIjC%STUjH?JWA6@pbZI;>D+7Ll49A3z;fstS9f%;5h-;txcezDpDI+El!#htB)a( zFRHIc#oZ|36Hk^~sNiB%f76URdRjQ>a$s)b`k{T<80+vWCE zIp_XjgO|r%-#n;pd^a*_*-DpyyNudZn zA-&n)A-XS&&7A54y-n9L22n89C?XBOLm{*@PifZS>h|t+*+&}N=FLa2-FX&u$>q_} zIYm`Bh-r6i0TFXwf04a(E$dN6EYD^Db}UKz--MR2A=Z@4yOzNyFEd% zRisN~(JE5C1PYUeQg_l_4ujZ#G0ruUhmnP4s0eK?-xeusOM=C@^UgQ}*Re^Hk`b~d zgUFgEB$we>x8-S25>=uPP!!~a2Q&u}757m*O>bpTwvRS+vDw$^#&}l}(SxD_hLQ`H z(h@MLvK_JfmUC63bjC47N#cafQS_*#u7jnBPC+p_nY^*ZcU5@XZI85nC3AU!#2@@EpmUAS7CcZsU=%h5OvukY}Ll;t=jKUK(Wjv^tns&ff7PS(K0c6% z93Me$v?E4}tc&Jqi5(eBY9@HtTun~2u@ye&^)%N!pNmUr(Jmd6A-^ed2o z1_qs}Ms35;(EG7};x^;wR<8w0GGfkcM|0VD+)zAn?_|Bz;`f{&)n`aO$CU+dnLAR<`%Z#?qLh=U{vr0YBxc#@+O$U$NcVl2VNcB zurfNekHPgSm!1Wb-X47fS^mLvEP4ZKcdy&)e{ps%;A$a%DHCmdzzp3(V)BRV`u}#o zwXgKNJI>4VH=yd^d`}&!ABC0Baw_lO1fnnOl#A=NeS$?x4=(h$pAn*is2kaX54vNG zgS(t3(yUM`$M?tY|7R=5b?LkT*{}RXB4%4l!5)2tvyI4ITgLT?|C2eN6k;|;>E}a? z^Nth8=Z-sn=mSx>!|8d)`Gi`)j`W1Ba-+A8+YYMSc7cw?@CreBuJAQ-7KQ1Iy{0XN^$ernjSyVoTU-JbP)(9QR4)Wb%d zpOT{+`YuGHRuB3GNAUZfFxmI)3M=NF{}Rbn$b%bc7)rM9PhV5j(?z;gPQb2<#^AnTaFLwmE)S| zs8a)fQJpi#9_1q1matZYHxXm#*ci&+WR_r|0k9I_i*s&HOjA5#(sA73ut$c#=oW19 zF0QeO>s{d*rMkj@%!Ylx+n>39f5zE>4|OIQ%f%hleb&bHp0$??>M?M1YQ*b*Jbv!H-34`g1PtKAPd{Jb(ck4ss)2gl zLC4{U=sFzUSy9-m{V5;$QK`V^AH85S=?{Fb3rBtAje}vpy8g)ZC*AIN>Uo2a*B!I| zps6E2Z0C`8t}H-rE}bgPwkp}xk3@NIiWh^;EbUf}c-tCRO$zv9arS{icr2s0~Nw9Y*)KI(x|obB=YKw@*~*uHAVys=0UjTZH%|>;zB9bQ2Jt zfQREm-fStlYP*d6l<+7Z`@t|e=YI!(_Wb9qW_50Ye_n@J)M{(p?oRlJqP*A3i9x-%!Wu4urTE=QlMB!t-)j7%ETac)trmSNdDfBxP7B#%P2Us?| zs7v@fLoZ{Y-Y3ho9#fPO4GeE?inWE9Fuyx7mliy}l?>Wf$-Yv(HWrr)ee_l*HqtS% zlpIFuj*X+Y6jU!@>QGv$lZSHU!KCO0!lYlmWQ$%Nkou_M zQ(-=T=>0Q8GI{?LXMXRanC{QnuiQ6w+eyBQ%J$Es-^Hw9`rO|fI0KCs!8!HY_rG=K zi#_XLa1maDTl1@VkgE1IdFH1JoZM$vS{l`VN6mSZch#t(Mk_lCHQ9=tPVm~?dlU7& zCdL;{XInUd^?`D^*U7D32^@32?it_WuNgdTErFx2s~24!3bXFg3|9}z`Es`VVyrU* zcA!<^v?pK3Or`ipUK9{p{w~)Y%TSf;?GHNiSmNd)zBsDoA5k297~Cq%21WE2vv2Jg zYkc()o(aY8JMeZTl}pH#*}1SJ&vF%Y6`pC%Hn?qQ?66`s8_0S#Zr%;cY=~O3SH4~? zUQ#jgdO^?C3;EJ-iMimSmjutva>W{3DPt*2c->+Xm?`c#Q3!m+fHfb@fS+kCfa?&8 zzv6w1=buVvR!ywvOxsMEOR@Fk$q&FgE9RY!3yp!vN-vDeW zfA|SE3P(i$65T&+VkI5#iyM79#<8*cs7lt|r@{qTjtsGqz||q8oIfoX0Q1j`p`5+7 z*~^a|1@1pH3r0eLZ*gC7kKaG-Yt(Nuy-t=Fkz5t}BWMYGt=ND&!brlKWXz(EW|n_+ zr$KZn1;BSS)KNI`o}Lx|eAXn8s1TeHgOU-Wf7ZAv(p;I+E4Eg7+wO%X2l=*aB-pU- zOVQX9gUuhDG_X%j(XIo4SB%8=D~2K<`aP(m3IVE~;n?WY+*fFvr2Ks7iDEo8r{Qwa z4}m+<36F2k#WEd2E`N#-8PH*{1fm6d!N{%=z^`y6P)+mj*yt0E;tQNvY0_FnA7GJ8 zB1H1T8*~?!8^7h>9$n4-8mb(Gog8fLy(X(?v>$u&v4UfuG#Kn;SK(+=SmIN z=WIQCS{3KCH0nB)grxdKKK=eUmg~uiwyY5J-jW>^&X#WZ<;53^uBM$8B&=4p^254w|u7+7%mo z-HaAgNIQVcqJ?|wY|}i(3#RntWNZ#}6jZ3}O7IqN;h}nYXwdBK$0JjDH5}HDR7S#_)r4$>!ep%&WK|t) z5pnskA!WF*9#6COy}v!hG+BCBiz08?92mwjoChw?xR^c|L-xa_sONraO-TU&IUBls zIAldIS`CVrzMGIVy>!1rV<TC}d21i0oc_?}tS-bjcs@U*HI zO=liIk${to+ZQ)!$o^<~eZaA@ezu^q#Oa}7$py9jo()Hy~;bRDb=dmJv~m{ z7BL>2cctZQh~v85yQWd~3waWcJtpT1A6Vxc0ak_Kn0m=_Yq~6?0xt;^-bnO+)SA__ z0A=fTrat*)NKKvxw<`5*(7hwvlreTuav`}crTMiZ69%c7C>m4K$U(Pmi7FzgI$NCP zpYu-cEG6Lyf|-bso12=FCh(X=GICZJEk;%)SogR{r~Q z48qzIv(7(%!$+#eIok_(wd^Gay5M}5G~43yF_)S){~@R^rQ@@SPf-{0TuYM^3!ZwK zV}x-0dqTeV)RRkLQgNgby55-i-p5+{s;Epbn|s527tf1is_0$1RD{%0cFLe{1G|}Neu3Ura_!+zWg0RE zKYDBVXLLUOq={@%HEuele4+`PS#wQtZJtACJ$F$uKU5uDZCp4a5^SiI9n^TEP}94- z#_{L44e*@6qjyu+|0tE`ITuxC;Mqr6G&LoEj&{fs;_zj;xktj`mWlUxwBzsB_5O&k zIB$4EE~uNY_48QI0XosTTtR@PpO~Qnz3lwTB+WNrWD&lJU8{@-lchqEDh{9NnV}_^ zRFmm2l2mjx5ywhC+fwC1K^<(0Txuj(B=6k|2f(|rP5q8QrYDo-6uwd8h<(N*RV9yn zSu<6IwOKST^Nc~Yq^k^Kdc<^d^wihw?DD)J3+}>AXT;VKx37FBkKRrE#Z1warU6C~ zuUBR1uJH7OIs8h}?>M#4A*r0an%i70KW`P^;FLYO;DQ#VAqYLO9>V&onQ3#-^<3ba)8@t~8E{K!aP(R0zW<9Dvf z6cPQCY+A7~{`E9wJlqt-vXG5vvE)h;jyh|%y>O*7XamkFxyRPd|6is7Kt12*e7 zJ|5PswC?0gG*viNSJ7__VE4aPs@^M$Gh^Q1rX>whw8|YI=w1iQd!;7S(xy+D_F-R` zZdXRQqDx?5&wH9j4I4cB;>1Hw$R;GE-JV=|Q2lf_PZ{g7Fe(zu;_>CRBvbHQ#Z;L+Qw+E;5 zUgS4zXXwGzlE2G`kM$dK!!@GB7-=@kv%wYnge{^ID9N!!IUAeIs&CN74Z5;D%qus< zGxrk7I=us16z5-@8iAY}8hIv⤅sOgcrwqXBHxr*j8(wI-y{-Q&9&01NI1Sn*gkY~L4ynXH=pNn!j$;lQNT5OsbB^0vkrAA0SHNk1;3 zCsn{9d-k+xh*MMDYndKUW{9a-s|2Z&m+$QtVIoj%$jcZ)zXMdH26Y|k$7+jp&yoL~Up=MkOnbw0T~l~%Cw zIRA3P)_b16go9(Z)K5oQM$@j~9E1-JPwGxn&S6s_8@D@!)s!g9jm7QD8OJi1_=iR` ztofe%_3{m9SoAN{?VcLxfM%r(fic_<2f}>kW5^Xex;;wOlwGtKbC*_ zBEl5A{Rj9P;=2FgLkl4c^k`sT<2xbY(Vx8IpG;|V!Z z&{U*g_xZb9Dmi;mF}2|EO?P%^mc%dQa1)%z0swAil5ro^n)3@c{ER7lmdVZl=scdwzn^fP4`=$Tca#Pbq#%NfO(Y4- znOGTVRN87Z;6~@&VZGwoRpM1G{=f$!gnR4D=HRA5sRUtSCK{=3Q=icNpa{1f4RctR z6!XI0OqVwp)0Oo&%dl;O(B647@g$)6{|)Ae<^0)hjAlU%{LO^9 zlmDAigP@jc9zvr+N0Vx!!MQBOVDZm@Q7n~jt{zbtZ@M7x_ta<3VfIhnzF;Zz4H`M9 z`yyPMUok*$s&2D|SwdG2Hxe2A4;47zr>{AxkUtPZ<9jFrmvV~LNWM07mhatOBX znQL zRnx_PC>00HC#n8eieD@KW5K^6C#L>QviR~lt#NidiNjrBf{MmKH%6i>>lH1j)=-vg zB71jGJ6FsCA5?o_C_F)%n}@i3BWkh!czc1mjEVbc!fh)dYoBm1HtVgR+rmyp4_ zeZ`Z?4%-3O4_H|UIIJdt-D$Iq(hUw1`VuQ3=UUzfz^ZGUt06$F?$+#IY}tl*=lB-v z80+Mt>@T?sGY_aJ3j%XFCRx{9{1qvtkCqAZ?^fK&Wtc zI%GqDN}-+nZ(Zu=X=Q%w4VER;BYV(~M|`)I^`W^?JwFHNlmy~wLqBFhtVwmxQDBP)m#;;44Kaz>euoZ zs=rc>*9n$z(3Ss6_(lEt(?vtdb+Y9_McG{~tG>pBFWv$oP9J6CH=Dg8JrbBJrv{o{jqjZfe73c1e0ht2yaAcHhm|+|urF)=ISI4be#a z9ehR5#^$4$^_%UWv{l7wyk`9dz(KI@VjB^ay}<4a!t3{Ud5&(qmie)Nmr#>5<0!F* z6z(}~PvhI2*A?t%Msqjy?CpN7_t8hzUL^1mTwIfQbv;T|ZI-&9DzQ61!Q{>CITMb& z_PmlBBGF?4n-9wDh`$Sn8D2{d=E}@AA0YL}5UQmP-P5zo70A}&`w|HQI#V_&LccY# zRv1A}u@*rdgZ>M`Zc8!9+HGWO1xxPw{N&$M@(M{Jj`rKr0@Af z;_)hYVY$6qoV6s8EZh2fJs`(&{I+XgAqIs{Zs9!bBWu4JgyH^5L3YR{_T$+?I(L^7 zZ8=3JWl`_*RA=F)uJe76L^F|jlWV0uPULMI=mb;14 zRDpUnayaT;Yd;+o2%#1Q!wVIV^L8AYVw>a#j16WzH%qoqZb8WGYT$E`7-I&if!(bd zg;xFx>S8rtit3=Moi43X8u*TJ4ZACXY7d=RTL>>4M8v+4qlyqKs}?}5KTSPZ=0jD1 z86c4ve<|>2yI744Kl2^zmEvKYmUMY}k%XI8B{M z{^vVM+*U#F`jp~NeKGQ!C;6?!KlJ(j$8ZJN^-sYW^JOLfZ9(A!L-2vbmqKzWq?bZ= zDdd+zaVeCSLREziq!zvS?OeqdNONxTTMOZzrK#edrL{}_jnMH(5cH2Sj08#iH(4y2 z#TRt{PoKIk2uxz38}F^lDfnECn1qPv|Hx+btDNM9D@lonrn!lTIEenU;SA#`jkxIl zUjTaUG8yqSn&1n%^(Xl)zf0VXmiGg3{=ufdaSqzs@8AB9zwtV{)(>=(>vG2b)LCL8 zBJ#^Yum0zFM7!wYOQ%0c*n!KSHnbD~;`&pZ|Lk$JIRIiJ(RGSUKtMVXPeJ4QtZ-!4B7kq1 z6U!A(W?2Z+())%NrV9phh5`93q8p|XRxFI2-!c}ZVn#Ej<;Ykllo?(Y!BWET;U!E2 zer1<`?A-Ygo&t;2u|sgUFGD$BI|~LwghiDr7#`dlR~>#Y0wHm&5~gqxG@eM}n8LLk zJq+GsNg{ZH!vl3aSvwx=jL%@i$_a#0pcLR8;n9(^66r!jw8S&P!=?oKukODea5Zk3 zfE9o(;mLx<@E7P(_*uvjoGDxdD{#%rqoS9O9m`AQ+^`iw4=!#EF&1h+Cfbmp6{tsTv-|o{2m4)km8lk5+9GMoHay$ z1AqMfe;=Z~a#%f>@fN$bG2zmGm60c)a0>i@N5Tv;9 zKu}qE&=kE)4xyHg1M|@OM{|yUq{GC=3^idVNZ2er=owAPk6eBe`Jw{pfl`_Z)Dr{b z&4s)})f$PRvAvx(#iNaKqhMO06Egp^7DA|W`k7KBhG_nH6~Bo?0U8I8HEQiF!He}d z1Em$14!*bNa>dqK^Ik}a90eG;J@E8>%G76LjueN9RgNC7tia_QmZU#-iUzAaRzae2E4aE$mFfLL+0`y`+XGkrs+qBTq6a zG$^gkJds&kgA(`vAqd*>5-q&u7TWE&y`~NUt~VKE9+;?$B?D6=U2G8wAVXywOI&}g zmY^0-h9mA0{A5|y@;xzspMzkSC9p?i-T<)|mZkXx2|f@FJW$pa*?epVpS$94KH;xD zT`+b$3~&=`Ji&R<#&=*$5@Wuj{8J@;$pi7UM4NLd(U5>S%oEbiV&h0q__Y~}2j?}w zNuUZ!BQ*AIWBc6ixPzwlS)`;SB%ob@aL%o}gkKB3+O7%YgpG@T4@%BLf+Ro5EqK^! zp`;G0oofAMDM52!$o*!CTslfD<1%I0eFuw};z^Oucd8-z!wc&g?j{qUpA`2I{R6kD z6fdcKt~o4&tp;s5)|_LhczDe_EQoUg$rEZ~@{(kJe8mTO5?VEh_L z1M!(D?PgUV9JMSJ^F>-#vOJDNlNmYBBDFGxwZ1avf1Rq6bBBj7A)oIB| z1%g6mMI2hogE(Ypc%>`!d*i1e$kB|=Z;6!L(`O>S`uK%wBLhLFyvY68pk6LW8Kdzb zndOFa*9B83<7%?BlE87oZ99d^6j;)K8vP4bH~{!Fy@0Xy`ZRRVsPNJAGn2+(hPPXFe2Jb&KfN_M zpX0IGrcj!6Sa=yf7ZGp}PA@_IBnQNkJ`ON9DiFS)>AKGxJ=qU7r5Aq0o>OLx@JlaW zq0=o(!~?R0|N0krg&D{CMRX0%#09}O{`>1DKvvizNAyQKUp)aZ%iT9|wOU=0K*P;y!miu!x6`6LGj~NNR#$UuioQw39Uk zuqXvXD~qMblyT;jz3p_?g@^C7T^UJ?z4KshQpzZE%PNi`dauqOa|vHGw1g*Ud7nCe zdL_$|z*P<>MQ%W`hTMG@nFEL`}E)J$?B zbv7xehcld%9?gT(k?!X%6z9A#cOfQk8ppu7b5h@!nGyki-q4`)Y`|)S2K}2lEd_H) z@kCtQs@$kdEp&g%2dphwx^8q^78%Vr=Q<{{Tm!N z;BKeoj4rgYORj8RN*bvhBBk6g{X6HS$#zVHwwY|cZgs@1FrO%b3k!n03C7fpcnO#t zW~Ir$Us~o;Hl5wrj!r|>*T;$Uz~n;984D%X1?fn53yNpB)Vq%vRe${cAE#TbI%|%g z4pkT<#+LE*pTt@=yX=>Kx3FJ-w)yYAWXJpn8Xb(acYq7rrR+@e04-Lw`q#eff4_HP z|0_y>t0u!J&X-EZl0l)+US;iYr{6_sk<7O#t4>e}QKz?MmXOQc@aSAg)CK%oC4U zQCU;EaN|&(Qd%Y`sjPJMEX;0raj6aH!jU$1#A~!IbTAlpE+MD4&f(CJE9J|WYsVuz z)C3f^Rw~$>7StIj>Q{t+_n3{BnQ@)b!%cQ^kwRElPL+tAa}7F&%oHWL!8kF9QB|~S zcA=NuU7HHrgPiANv^keB`iDh&D$$B+!BI&g%j5d$qMY2A$3~Yd;M&2=b|pe@q(5W5 zd?{U^BH>K(fX`yGgmf;85ir^)fB_@@Y$e_jE$pHKVqJAA*G=Ys57=p;0myD-B$vY1 zxt6EevTxr^?&l21?S_UHQa-rU`F``l{A@DzD`b1XSWY|U$jpJ`WkK#6Xy@kdN*@A4 z9<_L7G<(cnPlbP4@=IAu@V2#Yk^x~2PXfs)pP}FgSoUF*0K-Zuil-!{QY`uRr`sbx zTSFFuLgj~HodxoLY{kzs3przF$UyF@_V*7_`_El0?R0cXRAKnSM=9&)Npgl)9BN6zg;qe52y@%#T}2LCv28XX~Rh@MOI2bz>+%%@fgb&Orau<|&s zn$&^^)i~{1+(*QP^tBS;%uy&6!|>sLuUagCG(qQ|Z^zev3hM}@q~euzw5#W>9jrIQ zdhj`>GJmvdPZ_ebF<2-lqq9QH+TRv=vzYk_H1G2yMGeN#eRJuY2USxYT5Nz5VCb@4 zk;+7=#z7Y9r0b9+z(~w;X$8$L2f4@!`V0h5i6=8uQ!*es3@f6Y{eOVl(Tz(?WCbKLpw$2A}81x zPhzH3&IHIDsy!X~+kbE4cIa)vY6T)>;isr#`KZa7N43@EMyVV7yb;yXh4oqmAQ7%d zm6QyG$B|`Cljwdws;#KpsP)2-E;A*&a3>v9sa6U+o_^Sr?iU_b?Br06*nqX;u;BI5*-8%?x(E}qsfrnlY$(eSd=maPF87wQ^!1dRil%#Nd-?kDNiX4 zNgI8%gt1f_I|XATiWeTtv_Yr=y+`?fKIp_LOp*#oeJKadbi73Reg^tw?;sec1BE8~ zJ4BQrT>@Lc(#7S0xh4-=1CwsYqdAypK+vd`RMrCKe^A{%VquWhU6jf*C(Z~Z$hfQt zL1b1D2fBn)nHtVKDTwJ%C|-S%MW@C|q_MRy+wOBDS0U=1x17{Raz2)Nt!SBlsmAuLhb!mm$!7)3RF{Gctg5U$E#J zR0T{3ijeikUx>4m?8V_edvRKS*o*%NdS1Uf2nH@21zkRMr^DW;%X|L7bq7J$>-7ht zNuQ6Ku^;=+2U8(!YJ;itan4cD{@LWWny3`&N?*;W{6wm6p72Mp{LEk+73p=klafxY zadczIjiQIzL68!3DcJhIWxJBOmK^K9g<8UnE6$x4^2TvN2Y*+5o>nq{eXG{#sYw=~ z()V0nH?e)Dq|P++EXG+0^_5XUiB^FuUh9#>Jagz-6XneHc~gTp%Z#r`*m7#`8=nD4dX> z1U06ik(AWso;LGdD3fr1PZkF&Tg-dyyv95O!%puYFcMSDYvM?S;fW!)4V)|9!u)G2 zt!zmQv>b#lnV}&|5D5&gBbX03h9Gj5SPp!G4ia=J#(7+WvcgVFku1XP3=(l|C1FM7 z(4@B9BKjTp4+raofBjSO8q2&Kr+B};*tT-X*RH6oJp2XWZ77(3jro|q2>C3{tKi5< zf_G8>vKb_1Z|L3_6U9o>@o@5b`0S#%6D?Wk36J4sKY0z|? z#);W~p?UbmWAwA%lBNc~c1MHlzc;_On~6WG+#~r=9k~yNz*P()vHJX^N)0}~p%+HB z`8c%7#NEN|5mje@y!DEU{AL%$?GFa8u{DagyVY7pT4pu_73jNX6I=DTt)B8vrQ6Xz zUoKsChrNLW)en~;Q(K`D-wXJ-QN4KARuSZy41)(J7`4wzf2-?l zG~vfwVhmIR)6Emki&Uf;oKB(L6}Q~el_NJRB}VFqlSz(CK#vANaiRlIWelu}!HO8L zDG6%RbVnx@qnm^rlR)(sVy3;rb`)hZhDX;Tk7G?jY&DCaExfx^n|+5eoGB}R!zX$S@3&5Aj>Vjd^)CsvaL>AKo#g&a@5G{7(Fw zCC&+dyo-&(pcfxAAK?GF%+x6mFSc<)_qH&J!ui(d4J(8E6=^R<+S?s_BvEvQh_sDS zOJnKm?|GJDqT?sR$$NB?d|4H2Uxh&!;0oPduNC?qN&0r^esl~qj_n3#gcp98q00UI zi`x`sOLM5g>o9j08s8Iq`8Kd0y4|j8w=e&H)oysZyKVBnVd@n%8B9<&!jz+#ps31n z4IIUDPTtbgILvh|v%2ayF**1%znda!YQQlMRV^6`e;DRL#g00;^a!_tTGL=|7*910 zX)Dek=un_OiO0#^KY((mohL?zwHts8GM$&;l)%v$B9=ELoq+BTHG<8lpO$Vg(t1LF zyfj1@uv8|_TP+2&z42Qn5(M2LE?kbk)AD}(E&dEvXG43JMPOVRZOm}Lrl_PoRZBIf zY5^U4A_x$ugAA(SF%WW2#<~e0FxC|>#U3lx&x|~9Ap)G_+O?tn8DTKnatcEZ&PmY$ z!R9xS4IXetS-sb$5DQcO@Q#PtmtRhQ()DuoiV#ANrwQ)TW;!*nxMZ;R{uagg^i;`I#TcI3*$ zX(uvD8YeV={%UfY)y!rZY0DOU-i2iUsVta z-R%g4*f663bV4ksTpN&*71Kd~tk|IG=9x#2Q(O5Ue79q5*m5{j>#m~HqAITAR3H$F zpMhDSn`|h32c5pFvyUm7fd2k!aadg7B^M8%T0p9VHf7a1OogG56_S%!q9%sLEYh(@5vxHksJ;fQv^ zPq^hlHf;q3Nk_v(B1|iRhvo3C9g~}3kSI))TpI%?l|XsEZwAKEZl9J;TN&Lcm8ccV zX$k#^))N_3XwXjMyyMrGQZr6BU^h)YO6hR4hTWY}3;R}>)YZ~|Op7v*$OMw3Jyy25g!3P-3SC0l-57F~^nsQQgz%Gd&++aOkr&E+8Im=3A+ zNl_XKyp8&zQ)^s*nJXF9I_Mcl`#{^K2)Y?#16-}pQtAzCY9dklD5uWq#~^6D5x1L3 z$*eSrQhRkQg~n{{9p%rpy2EwNC1{Zg2CAl(J=N~Mq-L24!jy%d^ZMFW@TS$GCB^fY zWV-3X%J&nM#-c}br7N;_1yP|~x$kCAMuE_vT2+{c!tcMY3tYlO>fUA$4vcfIxE#BGq5~ISn10|1CzqeJ6V4Sk+jHAW zuL{~owSgn6%S=5m9Gc3Qa?h+MUnWi269f~f7s$|wUYs2cQ|%?5i|aDx9JbGUtd%N* zpqULlY_TkQNfm0sfJX|IdwTH!9`_ldzcK4j)5Kf`hbuR=%Vn-!FB6)0aQmrydM39m zRC79ib#dx)T)~j}aH67hUIte5L9Nt8*~3?yNs}~Db-|S+CPNxb@*>XSRxsbJ)I2o5 zV`B#GLCAYWB7u%|3aU6I5^E8SG{$)nUSZ7~QLxd<+ER+Rfp)kZ^%AKBV>4SCl9cYY z**?tnE_-*2UP<-cLR1?PCx=>#asYTks1=NV6&=voXkAyZTuOenc$%$HYVuCftx#z& zo2iI!A+&vYN2?t4Lx`x=gK;3?A|K`O*aUx5$scQO)DBJf z8CKie%6$s*ifN0u{$OEN=119aw5I~IGHH^3kb#n=$-<7CCW5uwQwuH!< zqr!&`xfU-(D3y#CYin)gmH|$G1~Xnyh|N+1VvP9^`}rJ6|MMA0RE-p&bc9s9WzvOI zr8H9Xb76=GLDv|zzGUGVsPaOIrRlP*{*1v_bxzt8{CUCs&m7&M9jDjr4Nf)0Hi4dK zSO-(aN~>L3=mZ%IHx--tG;#pgiphfX3=d6(?G7bWF1mrD4vsp9upWnhhMmC-MvzM5 z3DhSPra`V?!ewp83DdYjX*(Eq(bihG5wtcSx#FRA+_Zr_fK;Esc z(9;ai^r_Y*G%EkG39(_>Ufy!2u+%|C6Z1$bKqWFfEqXWHa>bS=vB!!2)!g0x%t#Z| zi-PoB?v`7itC;2pD9tg-lIw+Q(j_`he|(VF{40=??G{K-^;r^sOTFopx*D^}-#c%q zQnToet<51%%n@okZeEWe2^E&X#>JM#04 zfdd)EL&{G#YASPu&jz@pfu+?^AW_NPG(E-hJcyzPiupCl2hed;p1qOJ5L04Rwxl}A zJoh3`n4rKS*A3?CFN;mR(tg&eb`bXe*Gk3Gs6*D_)qrMy7Vs!YWHZ@ny|1!FnD`T7 zU9YP6_IwLx04j5cAa{n&yzL)}>BlCdhk5V9cei#asP~ zbgvnrA&@`8s8Ty~h>Rkq@nLE@5(zQ8)eT{3Rnk zU+Ho3r!NqX75UFW1XLB85--n>6n=w#Ud{atPq%3Sx%p9wmzxY!MERsXcOIA}GlL!w zmvejeI+=ASr0hf8 zj`8k9e{S7K-&YgvH}dS-I%r~QIm0@JA?x1dM>XWPJL*YVI^6%l*ILU(n{fAlFA7H3 zrqM^)1}1GwSNW$raG+u5KocqF`kD49xiNiz3{%>XJ=xOHvUwyi%j6iw`lXg&tsAf> z5O%lzk=myp1t-!SERJEj-ruUx$>A_{;!B>Dns-9U9e!$Cl>QhdeO&BnTS93|hx(0M z9U-?$4A$^5?ALo6ea;a%;69y7x`M}bVw=vnZ%1gEijW+3BOc=4Zt0dB7c<>xK6ea% zLspp=h-;yYck`HzYg6pC{*nIby4%`nw3~C=G^XpLAchL>&Xvdj@$J(E7fBfo%v2s6 z$N4-e26#-y?>9B_h}6hD)$bTKtmK&8#*gyWVJSr+oRf-Sv#p&_>t?tM%BZEoJtBSV z=#1RoGVHdHHl!#V6OW>VP=_x)9^e*#9>eBUIF+$)^GKu4LZx~cM%W5PJ-N-(K887d z<5f)!HI@WtLmKswaV=f)D|A8dF%0BnJs9 zbH6-bTY(Cq1lMxQoxIvC-i{#H(slX9HovjW<71m{@A&v8zd;NWL}OWnm>Pc!3w>DJ z_Jx$GvgiWX3@&O1h-Sg3i`6FvEdk}Whe2}meC`|A&elAFrrk4b+lREwlY*D{Ip!M?CIIixr zH`a(-co|{jzSb9ew8SC(o6x^0{i`nxBu<#n6ux@~Ux_s2_=09Tdgkc4eh^eBL_cEB z4W?M>}pa@$-udsdE|g2uPQJQd4-F%rDH;mf=OO+Z|oR=GK4x z3%q(B<8ARpbPdnM1;ID|`|Dzm0&yzT1unfjkE6L6>v-E(?T_`X7~cLT>3>E%!A`DQg(Ru8vHzxI5hp7Fmm8fO^3bVv(2w?Ns%?u8w4XV zZ!YjxW}Yg}SCh7X{h~rn<@b0zigR#LIv!QdS3ZGXIA!?fabZ((9I)rkD`F$mX~+W3 zQj9(&plQ>4EN_xQK0rUr6wE3|TJPljsL^K^YD$w z=x4tr5<+a1(O~=U3$dMVZRH-xhw8||+DWhCP^8u8Csk_j@r@DYTBzgDDie1Hw?|Z+ z@zyIY@|#_M6t_S409f{Lo6XQNvl*Q1@19L8!`a5l{&cv;;0}8OD-G=Hdi8~A8})^Y zzXNMg7b%O@aFUaCAeR@HG0>22Fl&Fyfbq)&SxtF?nh-EC3)W9m2deITmJ2rF@X~rzz=~fs!EFcNRXCAIiJ7Q!U z8e2{#)Y3FupF6)`lad@g-0lE@tQH-ZzdVh7ZI|I=e}Vd%xwEoQ(ol9r8;l4zPZDxj z-j6JQatH(-l|yE_yHNL-1P>`?1Nn&Ix}~WrPLYXZg^tOxN72lf21iOO!66gUo-9V1 zV~6f0T8p_go}dq>etVBei61C=S!-Ki|Vr@^zKM^kGHq!axi-oUGNJ?inYzCcKepZn!dTKrbZ0iimVwNfsDNV^@rYbDg!VLB;{q7B&Gi&UF9MLW0 zGhC>r5|42%*!mXn3Ye3{c@J+^jK{+aOxPc zj737em~E(S^$q{m{`9cMXvAqb;7HJ$2OS zXQMA`Q`9l&9!4E+s5@jCBpw9fLW1JNC~Azdp`uf+7z&4oHw1=0aGL=`@9)tyzD3F! zkCY9D)kvA`>h)A(WrKbjeP1SjR%RE!QP0Ff+b``G-fmOLd%9sE z>P=VfbLV}oZXr#u1a(3Ysg85iD}O1Mk+??i6Iu=YdQkqF#h~jEPX!)-&4K-4zL32X z7yoKlzB~|kJhohX$FSpI*g)9vezrRCY+tHv2;?;z>Ypm@E;c-A*2|@v{R9E6Dp0;o zA}5SE<(#Mtw_cebGWzNkYheF8qON>+?P*8%v4^e6bmwslDZ~6an+;b}0d_N%i&J2X zymHkxfpxXHZC_W!+clzp*1Lh`--(>1Xw1O+fNfB8y+ZH;IxNNR=zX9)nwhqu#$bgM4BkG>XDbis_t9UXZIC(CZYvcnnBli4+h z71U=R-}Ch$Z!kug4AiY07NV|b#NJ2Srt;{{f`ujDmf|ZiR?!K6gwAo}snJa<^JeO} zPq}Z&HcnmfymjUh&j!k&Gq@xSI*!j2Kg&BdmNxV09TnYs442B5qT%u}hdQyB7l{H6 zM61|flnzRz067j}S;?Q5FLf!$*_GOr9Mp9h~4;>Pu`+i z2wOK(%6ov6IIFs7%$hsbs{B)Q=ukKX53_=coCQ}R4;1P&d*>b4JcZkjcyr2CT5(px z_a@3^S%p3b96exIz^R|J?RJHVic`!lyM1TmP{pMF`29bBPT^i_bZLg_AzqsO-pGR* zJ|A*x?U6G_*mmc_Mz*xl^C1WIn*emqad`qGos46g8h!x||~6<}{(PI^b;t+f4DwlHc5TyR*ppvH1-UI=b;o z*Ov$f^}m(O!ax%zE7`PQuO*U!jtG`*HKn3*FNOC=;9Rlr+_{X9z7k@d$TJrSQ+dQ$ z+%ia3#tV+mhNqv12n-5RI4_D=D_9%4#Wr4{TW(u_wheYDI>>pV17D}wj9h z?V#s;jZUiygx2k?&C@#lQj#bocv|ycr+02~8H}D)UDBnm#v5N%yw%gfTMcIejc?ff zvxMV+g?H|F-aCNW2F}a;IgWYG7l}N|mWDfr*I^zDFvEi{-v;(Wx7*D}V}5nVQ=qH} zCM7=iI_ykNDyL1M&QIDmsNV7d6dsJDk=_%A=Q=^11e`u3PHc@au#1^U91pz}5sJbG zklwCV-gy9GQ_|Ja%zIz7slKCwi#Vo;S53-3pHjvMPau$$(QbvKX97+i%dAddJBdUV4rNtLIT`A9K z__8|m6(q$k3*_?UF(-`!Ly=m6 zbg7dib+-34P*#ZPN0{o3{{&L!Y%d9{;GS!;YmAFHurW)`%npxH56lnID3ALb= zTw&&4jz6C!oI7VWPC_YjaK{i#fD~_SB4fplP}UiyV{a@c_oZ5c3bAyw+F*iV+|(%0 zFCqgVR4vyaVf#p`1*V&mFm7n)t=HFR8MHL_l>7DalDC)*2nuLwoFe3Zj% z;bBVjMMMgVO&C8xkLUmVA3oykUdu$nF}$ge1-~=_JV<0#g{Syj%$M-AP!NJNPykyK%Y;E zV>utS*10X}AUGQIlZqUF^^ySJbu14JQnoIq$wwGcu?6lrtyyuSUwRznFb(beAoBR{ z4N>)3Q^MjfNpZ-nKauL4p+05G8j5`I?SUzs9)VH;H<&c4Ega3!Y&=sNl^*C36OAQJ zAI4U)NE2&2aA1jrpwnMO2)w9l)<=FE&z##F8kK=UNuu|KTqXg3d#lifGU_Okp_KK& zl4LV{fE#G5QIHtWv{Cw=cBd2JFCr>miPjibpm2kZ17L!lOlaJswC65CP&~rWkU?I+ zb8nMrmxVF9rvxvE_k_}^m2^LqB+*Gu( zlJ=S@syIRSpQ3_)%u*OqpT*m;Nc?#$uyWD|qZzd)tVlIuZ;Aso7f&lGdLyClobDK zHqX#|!GeD>%;c?EWgOSqKo|nRYh0O6VtD#)Fy9Aj=3D_v9klI1g zfb3w22-jQhKz=%vQnk8C)#Jf!1AVLNci)veKFzSZt=GT9bUYD<>)P4Uig&jWp&eg&G0Zw%4S>S%+2Np{}b%Tw3IViZtu6bCR!86cmk$a zIKD{yEQIrtYHP|96umKOR4EPh+_}_h#SOZ0dB?6T2~ATcFcYCc8Ou|?^CoSSh)lKA z(V1>Sz&FNa{qG&)G}U1N14v;Gdazln0e5}kX4?3Fo@cRMO1%VTP08mcfxPBtSak&N zPAy{jqV4~Td1P$famHYUma+orO)E;t&(n*9jr!qgSi_T{K6J)w8GVa zMmc)eTH)zHStyXgbjFGXBuFg*&!u)m(V5JDV+ln&7g zFf8ga1`X(KbNCW|OC5Z-c66>&wl4OB$^XT$sANRysfN}WFDpTqrs0=hvemd z!TUBg`-56%fVyn)=>qNx9lA80n3VxPL55?($M8i)kq$%~RKfhRQ0Ub&5o8nGkiGm&(!t~_7vKQ6Y zqvCFq@QEkOEmUx^s=sN*9X%}^bUCrVmDR0o1YSqHAqW+HM;q7w!s+@Cw zvBAq@uWueyH@+L0v}~nIz+J|2Ri%>70=iVGY#D2{UY0;QI@+=-!4cYsBb!Hml91kP z@DSY>#%4}+g5IWU8G|SoYZQ?N;Gqy&nx{1DaCLk4y6hv3ZS&?M*zP=wy5#cc=$xXe z8^pA`wt$GaufI$%Ss8XNdawm@zvuvi$qAW=rWEVaVPLjq5ChM71&31xfZd*;*ecQ` zvS<~lUIK+lL#aFIE{8$v80VUQ$-~IPGE{^%mv4&{wk5&h+<9l5f$P|$Ny!LVlR;$7 z6OzkttlRQ5D2Xc32Pg{i!ULLvh>H6to~E}lDBDLHy4dV%bz{6MiReL50Yk}!OKAxh zRoRYMe#^P4Q99$8q9k#`<|um9QrE#!M5myboJ`)>;=3xm?Y2kS61_5i+t7vfSem;O zk^X&EvoYwn1rm^{##eb&5ZJJAP0(V(*@CI@_%xHJJ8Gz8!xSDVgBci1bDl0ImK0FH zEY@vW^SSfZmMQAEevZ0K!faDJW4fS2m)vMBRw~dy9||24jYE@vj&7v(l4w4eZTP{h zrPVF$!s>8&IsOZzs8P@pQzQ5{GdLJLiM2?Rj zH`);+Mb<@gwZx8$B{dT~Y%cZt+&N*+rX;ZRSS-lV&?Ag|mJxP;1SqnJxM>wB8beCU zumXa^x9r_i>Km&u$ zRHL?GXz2Y|ahvgfbF0?^B^fd2wxhXhJZ>l+xp%VOYVmtckm@s}p5x3~!#4a*$B3hy z!kRLb4iib91REbLqLyUnl)VyLH**VHaQCnUcQ7jW0=1i-~K46CKAu;(wcKv@l;M!Mu z-W})V`5RDmaK5Jw)sMnTXgQU4a01a6cFM)|+CIUer3V*!+|LNnLDY@x!3W*3#=%|A z6KPhcmE-&4_y4n%*(#$)~-yS@ed*Z+%W>ZrOae~`-`hc5nX+JgW*WK$9h;Gk%KIrEAHtJ!c&QHnF z4Sg4)QL6|2f+P6-Pnhg`c7+x5&VPyID&)a@D_3oQ^<3oa=!X14$N&uOM2YlPRQb-p zosYy}FQX^})AS?17JP-;@@uUFcAVGO1_Ht~;~`8XhOo*F@#Dd$;|zN!(}#oUUFG;- zpB%56e`m22N~S-A(qmzMKz?Ac0z1NON{;v2Ajdz@%t1|hqH?_F)-A_}^~!NgbkwPV zsLq*xV~=tXZA(}y!kdUObZiXeZ!$};&;VEo@WnYdC#ET$G3hw&aM&Y5U~~&Mc^B8% z#PzOljZ$6VKW4+e-|f#_zdz$_zz4m_bTl0Mp4S~sdb8oI*XI+@4Vq#egElaazsV$E zk?c4h$mQaW>OO1Zde7R+1@#y>IyK^T9zS<~-tK}rJ^}{t;isRk@aXUIB-KE@?x5pv zM06bv@2n_n*8Y@_{HRpm^N(IIn)C<0*M*}#^2Wh1U|oOY`jc*VJoUW6$m@<-f6&yC zAGY(zJ69HPMoyH$@HP>)oEr8kX|a&u^dC%mb(O@<6wH?!tc;Zhtg? zJnIem!*jPczAYcOG_k&YbVmC^ucK0A;vZ8s9Cdp^e>xub81MGR-efZLW-}!W=?})BZTE+xJ4)WsK3Dr_!_}_R zWF5%mN6CUEgClS8N3OJ))TE(4ZUgtY()0gjso$Mx)a-G;bP`Q`PAXR{2~?%w#6D~N zg!ze7N`nmOWDl|Ax6+>4&ReSObU!65QkG>J?8J|94c}ZzSc-J*ym)T3eY*aC3O+m# z^5q}~G#+&vq@r&3&er~DpO0{~)kk=>j$oo7GO$y(;;DLRnkBDM8E%RLP1?YLp1P~B zcK5}oW>;ZOd*uE@;?vzFECo_Ap~~TLfXGBZ-8@Y%iIPoLwD&wVCIo6tqkhBwwBroA z=r$a5CwJu|`@Dvu$yPE!iLuFl}?=TawZx~r)*J~lq`j;9M*wUm&VHf<-l zeb7=S{R!}w-sELm2T&Bv)?JpIB`Z-tMFh!d35#TuEJ%5=4UJydarn$vOVW`@{R*-&Loo`}V!3@AOP})y#CAp5|Y_p&Sd$=#C#Qvryk{vTg~&-Uy>80TF}TQ40^bCM42w-3L~kFmGRMocof2XsuB+T9okur z=8Wm@k3Kek4rNqn(i+lv#`}2a?!gNcq9Z}OMYg56BTP@MJEmDW7KphhnX8uvRJIV~ zy_jmd-uGjYlv9~pZ4&EQttp=d8LD37;48D=E^%--O`gy&+~3pOn4L37yJHh0h1!OE zO1@o(SFmVVKiJC!pJt!b0BM8J!ayu+=HmVI?trUP#|b3d6o0U%~oqB%4`{XBgBEf0MCRR%1!w= zK!R>w(}X2!w5pn1YctW%*6^HX)1ua=&@aJh-vYegbB*CgpGEtwI5*j3+Q``I=PkY( z*~WWKp7BF~9bP~ZIq|3cr)Lj zCU85v8=qW-SmAa9SxYr@WA2kXQO&kexWFJ;R!`VBW}C6nj^)T9yoEDM*<(%8Esh?+ z?0a)%6XxQ@6F76fui@gaiZ;KH%O+X8f4oYHz%d|(ez=@z%Ik4&wEfXILx|=s_uB>- zow>16DIw*uu-<$dvvctJFziNDEO<@-Ia6me8rVZjWSu;m{ugM<0ZLrFnqyP(z z{-9yMT2`6S35BiYuOWPnawals8^)^TciLKoSq{vhtknZsALY012yjr_P3d{%ePww2 zUQh{|EpPqM6vp*4-_DZ%nIIWhU$xs**`a0Imvn@mrtLn9+{?8JJHk!+wZrMuZLNU0 zX}~;O5*Zvavhp6opuwr6+o8ag7JbKZmuY5P%PwtqJW2FJAp(4u*+yAvK0dtmj^EdT znh-sJR;yWRl2vE%6N?#cQhO9NG9g1Wxq;YQEDYeZT*n^@tZ$AD7K%)i{McIam52$q z7u6AWf<%P>YF*+bscceXT%?4~JIW7x0nG)4C#R+QF#K^yJeRA;t&dId>A%j~l4~sZ zJiRPu2WgylADb$3xA?lgu(FHIEGa!!5AGd9*c-RP(N$w#MTSSvqTqwbarQ*((|prT zw5M~;x+atW3wHZ?+L8Fqt2mR9BXwIe0zU*c?ctZBO$rIw8^T>i|I+5XXXxmdq#ck@Dg4P^B<}KKZErFjja-!u5K7am4Y!)SZsVd7 zaEM->%Dw#b*`!n1=RHrMP%FC_t9$p?+Nq(gr!gCgP=Y|1V|bajQIzmxyHMNZVZ(40 zR0*vW?{x#&v_kUd3Bzr#L;_3INdSeN4XN^MXQ(}TJz{gE$Tg!04A%fmR$IMN+HYse zxV`zQx#4-krtro0eu(0HQ@I5ZIa#}VSYyTPZrKCKz!vmecq-vh2Y^_ z4<|liS;)3BBX)xD@;Tos_DrnRQ!GTPe)yd+*k{2@95YIYxtEg|`b3sd{YByf#b^TB zfRWF{^JFq6Q4{x%Ixd8PH65OTq+TBGX{=V;RHw5P+g@e)>bs@$`wMowx+Ajc3pw=MW(ax$i&*~VoTXfZl@j}U6nOe3N?_$|Hk)F~G~5PHI%GuZo%ZOr z669E)SO1n3+mQv>qkzQknwv9nz5Nl91~;P?wQPpU8Vqfr7ezdY`|$N!{iSshgE;oZ z;CgD#_1x8`96lJ`pK_lvVy?3(!D}r%EGL%cG8P-cEpVCrJ%&gfYfmc|`5H(tX~k?2 z4{ZAj`|f&cd{xMXlN|2+K8vF|qj%j&o}eqN{-om2Y+u*WJQGxV!cx51gMY zb|`i~WxQE^%C&n7MSBD}Sek`bei57=4G6{MLTX7_t$lknzV6aCa}`E8zHEmaSTF5l?W5Z z5lUbqty#U^Nz&e{Yqq)bBS$sf236Ghi++wCFj}5{Yx;$dT|+Dut`;o8@{1X(Fz%*u zyPH_bu>WGxBSiXQLUQ_YO)n0!vhlxi>&vzD0mcKJ;GT>rt&xxTtFDzTiRjOco5Xt0 zR=k@F0gzz7Y=_a6dfIzh5o$+7Ju(^FzqOVdEf9j`gF_`St*Mjcac$`1fVFEoEA4~C z>t&gSW-vGPF&t*U&n!0Gdkew4_jo(qK>0nTnZWQWlTvAso+5@~m`84G@~Z<*r0D!L zz;})BK~N8E5p{&FLP&3BS424dSG^!WyZx#Sh<_+c~reU08epGnNb% zt(@{nWaF%N#zN3KXXOgb^`cFqEJD7L;5Mj+?JjAc>u3tgEr}sw^6U|(9sxHMydDj+ zP{fV_e0zLr$Q->6&TwpH8c@C`A!qpvxbymrZarD0Yk(zB5W-DH`_=$>he!Js#mw~q z-4K!jqxn%f!BetpYiCgD^JCJ$^v}*hn3}(a6FL68b|JhOv}dYUvO~KB zr)nC}d5xYSf<~Zat)4-IhhSaO)?x<=&nGuha0KPweD_LON6#x}=HO1S#ep7xmiK@qkdBMhqL8jH~Ha&4ZoGsl*+pdqh}{+GF1e3lUqbY--d$yx}Q8< z|AE}p)h+OSl<9dS!uarZn!c+l5aL`rW8*`x=9}Tbq;yCBpa=4E9crbH9=U(um+IfT z{)>%TRaAm9z`{)^UDID^*DmOLb=T!f-uDBRlAvXDUe>B(vtiNo^HXJr7QU)|cPI+Q z@{2J2%)~;$T+K&QTyVwLyLjm11NA~^fp$ePn61S`mQ#K@28#&i^PTnqdQYZyX8rnc zm@&!8iRith{eX$5w^NNf1>=2^BJo;sEBlv1_~PJMvOSNcB70rS7k%#u9zIpA=`xa3 zx%=G8xzwWGu9I0gI1HE0TFXt^Y`~vL)pumSWg;We$6Rf;sv p5_!03I3V+sQU>kF_1D@j1qq?e1m^;1@TY*BE(aGjPfEZBybL zw5>=gmU-8!I!rc&1Y|oz2=x$zae?Fo_+`(MJYO$s-%YWN2-S9X4qB{SBbtAp5Bq2u zi($~4#^JClfA%G0JNYRf%hqee?9D_`^u;38I`AR3T~j@nyMrBo_yX5KUJP3L^_A4z z?q10H;cPFE1mRIjBPH}5`wbsEK8JyVIbNO$nNU))?Lu4j5~4#j!5{-?tn8==OWkz?CWBCR4NKuF`w@ui$5x$p+$L7uTl5ly?(Nfe1af z@>VovN{sD!RG(6Lxpn$WbbTYuy&p^*#mIXd=WUx+M5XGlnncJ$9aIIgKAcOGWp)j& z?R=R>?ObYLgVQ3<}Df#tvz{5`JEYTBgl0YlvRO@RY zd-0(L-T`|byhgsJO$?J@Ite_RI@1Es>#ci9HmHI;e3P;?@v&IvzPRLNM;;OSyYhC) z_Ju-ATT@3}cH_vJv$Wv$BS!WD1$W;;qUPuH&%eJ5UegF!=XfkiPoibRe|n zwC~Pg)0K89W#*hYpC^{Xd=Hk@a0;~12Uh-I5cbz2LYCuhCZ6Ql|t6KFO!S?U;@#`K?{^pLb@S*@ty$j#3T!@IXio_mEnry(U z-mH(#C2TSwx}txOgtT99?;cpay)&iVk@w(;AKFNJnDV3e4ibSZ8kK3k+gT&?RB$Pc{;*H#AaabWsqpXQywd z)kxP0V^?7|EByXH`)Fx@^Z!z3@l=cr^!n_PQ9kACx9Xdr*A)M-GhAjan_8k(d|^WZ z@AXfNiR6brA;+A;PyOio?IE zU@hkVC8^kNHA!-1sdP~Km*v2b5Rde)$RaEP|9U=>(D@!9a~8|br_N%l_c`)q(ThGd z4TNpWB6WucJxB4<@()Gcq}QOa=W3qQL4G??Lq8gz-`I|GK91c|N#O}qH7?dfKMWBc zSy`ITD~sTZ+^gOjzEtDGdzMaG;GS-)HvgzCY&3U~GoFktzH~25zy25j5!kO+`ub}p zQIp%FGqws3NUeRFD$7}rnzX)5&HaEZig@tF-3l(g(W;a_?516-v48GhoHi9ES^I}c zFvlSO+6feE^iJ$pa1@ugAvsr6Eb_9NCUx)){cT=pt1~2~7@@C8{nY99?jci#S~0$hbr>zA^?17_pn}k20L@nLw*{Q7-2Y#!isNAemfp0f|6w1Xc zzdrT&+!%V(z%1194QS{FHu^W;{%c)1++0#3?K8xJ&8q&w!7BENSLR?>kblY|NBW-(+xqZd{hl?9UbJ5O2GBI>H>}*ZvU3>WdxZ72~T@wFw=j;KM!C<6=+kHSW z^_C{5djauy9e=qC<)`5{!=w&{7Y*t31dHEt*-nZo?J$p?HoP$!=`D@*ovBx?LsXeC zt64Fmj1OZEC*~Te6Urq^%Bf`B^PAOVF~>fW*L;i~{&|95x+{9_e54Us_U5~i5s&ec zr_GE{FkuC^bOz#;_p&M&%t@Fo;IJ>iJAl$8Pugy&NuFIAfi`I%kXL@9QES%T^mS(f z=bP1|5`wO0FiD$=Ss3%=?C%omcH5pfSwPicJOzr63o!yx96P#{h3~Si|*H2h$ z`u!iT%w@*Jqw&TS2oYTzrwL05ufKEM^0>A^6?mH26hsYWJ2?Ecz^Ct!_TsGxb1G3j z9b7=8Lh;6m90mDg9$R7GC;s`FZjir0hJ8Ek;ABkamb>3}T46X|_O%ggkdrJ3ZFa8}3 zCbii`1^?2|4My59v2(|rn~*LpzH8ZsCd4w7fiI_&H&C)2ST#X5PR!jhM}V)JVA)EL z`2PA#!EXwk$yAR~1sehv`_YNKxdkVP%A2L;XWut=XH^WBCz2dxNfN{=tI5MxA7&gc z-V8^x2x&7b*hf6dczk1@%H(=XMckpe#%cVY_D-&%QZyBurqo;K_WIFV{6!_opyBS< zXiRN4$3GuQTF-|>SKEt)-yiVb3Q{Ss>&ieA)bX$VW-|CX;a^MmL0I%b7~Unom%zUS z!6gVUL39b?OOT)s!bpWL{yL-i!pP3e|JI`UXUNd}GvxLee_N=OAPVSJWh96~;-6Zv zDCS@ak*hwPUuTVEL7@MY zw?*Y%_Tep|8bT;;UJ+hxyqAL#;)6gm|3(J_ZJb1^3OrQxQ0+ zE}edvRU_$tP6b{N@@5b{-~V0KWd@f4U8+tqD+D0WU(o*}qhF}TOV+EjH@80)gD*%y z7RqG*;k1TQKB!0-dCy{{%;+e+#4 - + Navigating food web prediction; assumptions, rationale, and methods @@ -107,12 +107,13 @@ - - + + - + + @@ -142,6 +143,7 @@ + @@ -295,7 +297,7 @@

Navigating food web prediction; assumptions, rationale, and me
Published
-

May 17, 2024

+

June 4, 2024

@@ -360,7 +362,7 @@

Table of contents

At the heart of modern biodiversity science are a set of concepts about biodiversity, community structure, productivity, and asynchrony, and how they define the stability, resilience, and dynamics of complex communities. The use of species interaction networks provides a powerful abstraction that one can use to help quantify, conceptualise, and understand these concepts. However, network ecology has its own nuance and idiosyncrasies that not only provide a barrier to entry but causes dissonance even within the field (Dormann 2023). This is perhaps particularly pervasive within the space of network prediction…

One of the fundamental challenges that we are faced when working with and studying interaction networks (and, within the context of this manuscript, specifically food webs) is that there is a scarcity of ‘real world’ interaction data (Hortal et al. 2015; Poisot et al. 2021). The difficulty of recording interactions in the field (Jordano 2016a, 2016b) has necessitated that researchers find and develop alternative means to construct and build food webs using models (Morales-Castilla et al. 2015; Strydom et al. 2021). Over the past decade, there has been a proliferation of tools and processes for characterising food webs, these models span a wide range of philosophies that rely on different approaches, data, and definitions, which ultimately determine how the food web is constructed and coded. Although the development of these different models have carved out the path for constructing either synthetic, ecologically plausible networks (Poisot, Gravel, et al. 2016), or providing ‘first draft’ networks that can be utilised in real world settings (Strydom et al. 2022) we are still lacking in discussions that are explicitly comparing and contrasting how the way one chooses to approach the task of constructing a food web is introducing (and ultimately embedding) specific assumptions and hypotheses (Petchey et al. 2008). Most attempts that focus on comparing and contrasting models are focused on the same group of model families (Williams and Martinez 2008; Pichler et al. 2020) and only benchmark the different models as opposed to contextualising them within the bigger framework of understanding the data needs of the different models, as well as how the resulting network is defined and structured. As food webs become a more integrated part of some of the broader fields of ecology (Bhatia et al. 2023; Thuiller et al. 2024) it is critical that we review these different model families as a whole (not only in isolation), and move away from simply benchmarking the performance of these different model families. This is important because different models impose different constraints upon themselves and will not only delimit and dictate the potential questions one will be able to ask (Petchey et al. 2011) but also determine the appropriate research setting for which the model (and resulting network) can be used. For example the use of ‘structural food webs’ are useful for developing additional theory such as re-wiring of networks (Staniczenko et al. 2010) but would be meaningless if one’s intention is to produce a location-specific network [do we need an e.g., ref??]. This will allow us to ensure the right models are being used to answer the right questions, particularly within the context of trying to accelerate cross-cutting research in the face of global change.

-

When navigating the seas of using and constructing food webs the researcher needs to be able to clearly articulate and define the parameters that are used to define their food web(s) of interest. This will aid them in being able to select the correct model to help them to reach their goal. In order to be able to make informed decisions it is important that one has a strong grasp of exactly what it means to ‘code’/define a food web Section 1, a clear understanding of why one wants to predict a food web Section 2, and ultimately one needs to be able to asses and evaluate which model family is going to best match up with the goal of network prediction Section 3. Here we specifically aim to look at not look at only the performance of the different models but also initiate a (thus far lacking) discussion around how the interplay between the language used to define networks and the underlying theory/philosophy should also be a part of the broader discussion when it comes to the task of ‘model selection’.

+

When navigating the seas of using and constructing food webs the researcher needs to be able to clearly articulate and define the parameters that are used to define their food web(s) of interest. This will aid them in being able to select the correct model to help them to reach their goal. In order to be able to make informed decisions it is important that one has a strong grasp of exactly what it means to ‘code’/define a food web (Section 1), a clear understanding of why one wants to predict a food web (Section 2), and ultimately one needs to be able to asses and evaluate which model family is going to best match up with the goal of network prediction (Section 3). Here we specifically aim to look at not look at only the performance of the different models but also initiate a (thus far lacking) discussion around how the interplay between the language used to define networks and the underlying theory/philosophy should also be a part of the broader discussion when it comes to the task of ‘model selection’.

@@ -887,14 +889,14 @@

References

and P. Beckerman, Andrew}, title = {Navigating Food Web Prediction; Assumptions, Rationale, and Methods}, - date = {2024-05-17}, + date = {2024-06-04}, langid = {en}, abstract = {TODO} }
For attribution, please cite this work as:
Strydom, Tanya, Jennifer A. Dunne, Timothée Poisot, and Andrew P. Beckerman. 2024. “Navigating Food Web Prediction; Assumptions, -Rationale, and Methods.” TREE (One Can Dream...). May 17, 2024. +Rationale, and Methods.” TREE (One Can Dream...). June 4, 2024.
-