From 617a42660908dd03a8ca700636c251598f14a208 Mon Sep 17 00:00:00 2001 From: Eduard Kerkhoven Date: Sun, 1 Nov 2026 16:20:18 +0100 Subject: [PATCH 1/3] docs: README with the documentation links and the RAVEN logo for dark mode --- README.md | 164 +++++++++++++-------------------------- img/raven-logo-white.png | Bin 0 -> 10765 bytes img/raven-logo.png | Bin 0 -> 10920 bytes 3 files changed, 54 insertions(+), 110 deletions(-) create mode 100644 img/raven-logo-white.png create mode 100644 img/raven-logo.png diff --git a/README.md b/README.md index e3b02a4..c29b685 100644 --- a/README.md +++ b/README.md @@ -1,78 +1,53 @@ -# raven-toolbox + + + RAVEN + [![CI](https://github.com/CompMetEng/raven-toolbox/actions/workflows/ci.yml/badge.svg)](https://github.com/CompMetEng/raven-toolbox/actions/workflows/ci.yml) -[![PyPI](https://img.shields.io/pypi/v/raven-toolbox.svg)](https://pypi.org/project/raven-toolbox/) - -**Reconstruction, Analysis and Visualisation of Metabolic Networks — in Python.** - -`raven-toolbox` is the Python counterpart of the -[RAVEN Toolbox](https://github.com/CompMetEng/RAVEN) (MATLAB). It builds on -[**cobrapy**](https://github.com/opencobra/cobrapy) for everything cobrapy already does -well — simulation, standard analyses, SBML I/O, model manipulation — and adds the -functionality that is unique to RAVEN: - -* **De novo reconstruction** from KEGG and protein homology (BLAST / DIAMOND). -* **Context-specific models** from omics data via **ftINIT**, with task-aware gap-filling, - the linear-merge MILP reduction, and metabolomics-informed reaction scoring. -* **Metabolic-task** validation (`check_tasks`, `find_task_essential_reactions`). -* **Gap-filling** — connectivity gap-filling against template models, plus LP/SWIFTCORE, - MILP, and topological strategies. -* **Omics integration** — Human Protein Atlas (proteomics + RNA-seq) ingestion. -* **Sub-cellular localisation** prediction by MILP, with partial-update mode and pluggable - evidence sources (DeepLoc 2, MULocDeep, COMPARTMENTS, UniProt, …). -* **N-model comparison**; **reporter metabolites**; **FSEOF**; **flux sampling** (ACHR, CHRR, - and the classic random-objective vertex method). -* **YAML I/O** following the cobra standard, plus geckopy's `ec-*` enzyme-constrained - fields, and **RAVEN-style Excel** export. - -The status of every RAVEN function (ported, mapped to a cobra equivalent, or -explicitly not ported) is documented function-by-function in -**[RAVEN 3 and raven-toolbox](https://raven-docs.readthedocs.io/en/latest/raven3-vs-raven-toolbox.html)**. - -## Design principle - -The canonical in-memory object is always a [`cobra.Model`](https://cobrapy.readthedocs.io). -There is no parallel RAVEN struct, no `ravenCobraWrapper`-style adapter. RAVEN-specific -fields that cobra doesn't model natively (`rxnMiriams`, `metDeltaG`, -`rxnConfidenceScores`, …) live in cobra's `annotation` / `notes` dictionaries. This -avoids duplicating cobra's data model and keeps raven-toolbox interoperable with the wider -COBRA ecosystem. +[![PyPI](https://img.shields.io/pypi/v/raven-toolbox.svg?include_prereleases)](https://pypi.org/project/raven-toolbox/) +[![Documentation](https://readthedocs.org/projects/raven-docs/badge/?version=latest)](https://raven-docs.readthedocs.io/en/latest/) +[![GitHub Discussions](https://img.shields.io/github/discussions-search?query=repo%3ACompMetEng%2Fraven-toolbox&label=GitHub%20Discussions)](https://github.com/CompMetEng/raven-toolbox/discussions) -## Status +**[Documentation](https://raven-docs.readthedocs.io/en/latest/)** · [Installation](https://raven-docs.readthedocs.io/en/latest/installation/python.html) · [Guide](https://raven-docs.readthedocs.io/en/latest/guide/index.html) · [API reference](https://raven-docs.readthedocs.io/en/latest/api/index.html) · [Compare with RAVEN](https://raven-docs.readthedocs.io/en/latest/raven3-vs-raven-toolbox.html) · [Citations](https://raven-docs.readthedocs.io/en/latest/references.html) -**raven-toolbox 3.0.0b1** is out — the first beta toward 3.0, tracking the upcoming MATLAB -RAVEN 3.0 beta ([release notes](https://github.com/CompMetEng/raven-toolbox/releases/tag/v3.0.0b1), -[changelog](CHANGELOG.md)). It has been validated against MATLAB RAVEN on **Human-GEM** (5 Hart2015 -cell-line models, Jaccard 0.975–0.980 — see -[the Human-GEM validation study](https://github.com/CompMetEng/raven-docs/blob/main/docs/parameter-tuning/studies/humangem-validation.md) -on raven-docs). +# raven-toolbox -Two deliberate scope decisions, not pending work: +Reconstruction, Analysis and Visualisation of Metabolic Networks in Python: the Python counterpart of the [RAVEN Toolbox](https://github.com/CompMetEng/RAVEN) (MATLAB). raven-toolbox builds on [cobrapy](https://github.com/opencobra/cobrapy) for simulation, standard analyses, SBML input and output, and model manipulation, and it adds the functionality that is specific to RAVEN: -* **Classic tINIT is not implemented — only ftINIT.** raven-toolbox is a new - implementation with no installed base of tINIT-built models to support; MATLAB RAVEN - keeps both algorithms for backwards compatibility, but ftINIT is the algorithm tINIT - was superseded by. -* **MetaCyc-based reconstruction** is not implemented and is flagged for removal from - MATLAB RAVEN as well — see [IMPROVEMENTS.md](IMPROVEMENTS.md) under `R-MetaCyc`. -* **Dynamic FBA** is not implemented — several maintained Python packages already cover - it ([`dfba`](https://pypi.org/project/dfba/), [`reframed`](https://pypi.org/project/reframed/), - [`mewpy`](https://pypi.org/project/mewpy/)). +- Reconstruction from KEGG and from protein homology (BLAST or DIAMOND). +- Context-specific models from omics data with ftINIT, with task-aware gap-filling, the linear-merge MILP reduction and scoring with metabolomics. +- Validation with metabolic tasks (`check_tasks`, `find_task_essential_reactions`). +- Gap-filling: connectivity gap-filling against template models, and LP, SWIFTCORE, MILP and topological strategies. +- Omics integration with Human Protein Atlas data (proteomics and RNA-seq). +- Prediction of sub-cellular localisation by MILP, with a partial-update mode and exchangeable evidence sources (DeepLoc 2, MULocDeep, COMPARTMENTS, UniProt). +- Comparison of several models, reporter metabolites, FSEOF, and flux sampling (ACHR, CHRR and the random-objective vertex method). +- YAML input and output that follows the cobra standard and adds the `ec-*` fields for enzyme-constrained models of [geckopy](https://github.com/CompMetEng/geckopy), and export to the RAVEN Excel format. -## Installation +[RAVEN 3 and raven-toolbox](https://raven-docs.readthedocs.io/en/latest/raven3-vs-raven-toolbox.html) gives the status of every RAVEN function: ported, mapped to a cobra equivalent, or not ported. -```bash -pip install raven-toolbox -``` +## Design + +The in-memory model is a [`cobra.Model`](https://cobrapy.readthedocs.io). raven-toolbox has no RAVEN struct and no adapter for one. Fields that cobra does not model, such as `rxnMiriams`, `metDeltaG` and `rxnConfidenceScores`, are kept in the `annotation` and `notes` dictionaries of cobra, so that raven-toolbox stays interoperable with the COBRA ecosystem. + +## Status -That installs the latest **stable** release. PyPI pre-releases are opt-in, so get the -3.0.0b1 beta with: +raven-toolbox 3.x tracks the RAVEN 3 pre-releases and is a pre-release itself ([releases](https://github.com/CompMetEng/raven-toolbox/releases), [changelog](CHANGELOG.md)). On Human-GEM, the five cell-line models of Hart2015 extracted with ftINIT have a Jaccard index of 0.975 to 0.980 against MATLAB RAVEN ([Human-GEM validation](https://raven-docs.readthedocs.io/en/latest/parameter-tuning/studies/humangem-validation.html)). + +Three functions of RAVEN are not implemented: + +- Classic tINIT. raven-toolbox has no installed base of tINIT models, and ftINIT supersedes tINIT. MATLAB RAVEN keeps both for backward compatibility. +- MetaCyc-based reconstruction, which is flagged for removal from MATLAB RAVEN as well ([IMPROVEMENTS.md](IMPROVEMENTS.md), `R-MetaCyc`). +- Dynamic FBA, which maintained Python packages cover ([`dfba`](https://pypi.org/project/dfba/), [`reframed`](https://pypi.org/project/reframed/) and [`mewpy`](https://pypi.org/project/mewpy/)). + +## Install + +The 3.x releases are pre-releases, so pip needs `--pre`. Without it, pip installs the older stable release 0.4.0. ```bash pip install --pre raven-toolbox ``` -To work against the unreleased code on `develop` instead, install from git: +Python 3.11 or newer is required. The unreleased code on `develop` installs from git: ```bash git clone https://github.com/CompMetEng/raven-toolbox @@ -80,67 +55,36 @@ cd raven-toolbox pip install -e ".[dev]" ``` -raven-toolbox requires Python ≥ 3.11. Genome-scale ftINIT MILPs currently require **Gurobi** -([details on solver portability](https://github.com/CompMetEng/raven-docs/blob/main/docs/parameter-tuning/studies/init-solver-benchmark.md) -on raven-docs); toy and unit-test work runs on the open-source GLPK. +Genome-scale ftINIT problems need Gurobi ([ftINIT on different solvers](https://raven-docs.readthedocs.io/en/latest/parameter-tuning/studies/init-solver-benchmark.html)), and small models and the unit tests run on the open-source GLPK. [Installation](https://raven-docs.readthedocs.io/en/latest/installation/python.html) describes the solver configuration. -### External command-line tools (BLAST, DIAMOND, HMMER, MAFFT, CD-HIT) +### External command-line tools -Some workflows call external tools. **For most users there is nothing to do** — -raven-toolbox downloads each tool it needs automatically the first time it's used. +BLAST, DIAMOND and HMMER (and MAFFT and CD-HIT for building HMM libraries) are downloaded the first time a workflow needs them. A tool on the `PATH` takes precedence over a download, `raven-toolbox-binaries --set runtime` fetches the tools in advance, and `RAVEN_PYTHON_AUTOFETCH=0` disables downloads for air-gapped systems. [Download data and binaries](https://raven-docs.readthedocs.io/en/latest/installation/data-and-binaries.html) lists the tools per workflow, the cache locations and the platforms. On Windows, homology reconstruction and the KEGG species model run natively, and building the KEGG HMM libraries needs WSL2. -Which tools a workflow uses: +[SuppOrth](https://github.com/fma91/SuppOrth) orthologs can start a homology reconstruction. `hits_from_supporth` reads an existing `suppOrth.tsv` and needs nothing else. `run_supporth` needs the `suppOrth` command, which is not downloaded automatically and runs on Linux and macOS: `pip install 'git+https://github.com/fma91/SuppOrth.git@stand_alone'`, then `suppOrth install`. -| Workflow | Tools | -|---|---| -| Homology-based reconstruction | `blastp` + `makeblastdb`, or `diamond` | -| KEGG HMM query (`get_kegg_model_for_organism`) | `hmmsearch` | -| Building the KEGG HMM libraries (maintainers) | `hmmbuild`, `mafft`, `cd-hit` | +## Quick start -Optional: +```python +from raven_toolbox.io import read_yaml_model -- **Fetch them up front** instead of on first use: - ```bash - raven-toolbox-binaries --set runtime # blastp, makeblastdb, diamond, hmmsearch - raven-toolbox-binaries --set build # hmmbuild, mafft, cd-hit - ``` -- **Use your own install** — if a tool is on your `PATH` it's used instead of a - download, e.g. `conda install -c bioconda blast diamond hmmer mafft cd-hit`. -- **Disable automatic downloads** (air-gapped / conda-only setups): set - `RAVEN_PYTHON_AUTOFETCH=0`. - -**SuppOrth (optional):** homology reconstruction can also start from the consensus -orthologs of [SuppOrth](https://github.com/fma91/SuppOrth). `hits_from_supporth` reads an -existing `suppOrth.tsv` and needs nothing extra. `run_supporth` needs the `suppOrth` -command, which is not downloaded automatically and runs on Linux and macOS: -`pip install 'git+https://github.com/fma91/SuppOrth.git@stand_alone'`, then `suppOrth install`. - -**Windows:** homology reconstruction and the KEGG species model work as-is. To -*build* the KEGG HMM libraries (needs MAFFT/CD-HIT), use **WSL2**. +model = read_yaml_model("yeast-GEM.yml") # a cobra.Model +model.objective = "r_2111" # growth +model.reactions.get_by_id("r_1714").lower_bound = -1.0 # glucose uptake +print(model.optimize().objective_value) # cobrapy +``` -## Documentation +[Getting started](https://raven-docs.readthedocs.io/en/latest/guide/getting-started.html) continues from here. -[**raven-docs**](https://raven-docs.readthedocs.io/) is the documentation, and it -covers MATLAB RAVEN and raven-toolbox side by side: the same task shown in both -languages, with the examples run in both languages by the documentation's CI. +## Documentation and support -This repository carries no prose documentation of its own. The API reference on the -site is generated from the docstrings here at build time, so a docstring edit -reaches the site without anything being copied across. Contributor and maintainer -notes are in [CONTRIBUTING.md](CONTRIBUTING.md); the roadmap, the backlog and the -MATLAB parity record are in -[raven-gecko-parity](https://github.com/CompMetEng/raven-gecko-parity). +[raven-docs](https://raven-docs.readthedocs.io/en/latest/) documents MATLAB RAVEN and raven-toolbox side by side: each task is shown in both languages, and the CI of the documentation runs the examples in both. This repository has no prose documentation of its own. The API reference on the site is generated from the docstrings here when the site is built. [CONTRIBUTING.md](CONTRIBUTING.md) has the notes for contributors and maintainers, and [raven-gecko-parity](https://github.com/CompMetEng/raven-gecko-parity) holds the roadmap, the backlog and the MATLAB parity record. Use [GitHub Discussions](https://github.com/CompMetEng/raven-toolbox/discussions) for questions. -## Relationship to MATLAB RAVEN +## Cite us -`raven-toolbox` is an independent Python reimplementation of the -[RAVEN Toolbox](https://github.com/CompMetEng/RAVEN), released under the permissive -**MIT** license. If you use it in scientific work, please cite the RAVEN 2 paper: +raven-toolbox is an independent Python reimplementation of RAVEN. If you use it in scientific work, please cite the RAVEN 2 paper: -> Wang H, Marcišauskas S, Sánchez BJ, Domenzain I, Hermansson D, Agren R, Nielsen J, -> Kerkhoven EJ. (2018) RAVEN 2.0: A versatile toolbox for metabolic network -> reconstruction and a case study on *Streptomyces coelicolor*. PLoS Comput Biol 14(10): -> e1006541. +> Wang H, Marcišauskas S, Sánchez BJ, Domenzain I, Hermansson D, Agren R, Nielsen J, Kerkhoven EJ. (2018) RAVEN 2.0: A versatile toolbox for metabolic network reconstruction and a case study on _Streptomyces coelicolor_. PLoS Comput Biol 14(10): e1006541. doi:[10.1371/journal.pcbi.1006541](https://doi.org/10.1371/journal.pcbi.1006541). ## License diff --git a/img/raven-logo-white.png b/img/raven-logo-white.png new file mode 100644 index 0000000000000000000000000000000000000000..ec34f826213d4b19cdac6574bafdce6b5682ec55 GIT binary patch literal 10765 zcmXXsbyySJ*K~JxBNEb`3PVIXq+>KlcgK)MKtz#721vsY=^CR{>K)yTjuDfR62IZ^ z`)AK~?>%`>+?{ih^>sB!iI|8mFfd3pUjPg-FtC)+-@g;!p+B(=^kXqF_@^}i&x}JB zj|$y#Se)7uPI5VGIB}>!*f@@~0J0jPXD#2U$Pze!u@pRq`2^p|VmV&Mem(haKOdYP zb66P@OSzy}OO=g{qm7k=%fr*JUCi!1Q}wbU{7WYiS>I4yblb6$S6A_+rl-2TVr@Hg z>18>UJ?A2b?6bsW6DIT&N&vJ(SVD=Qv(Qa>M^ysOxW7VQ)KGuFVe)=fxN^mp#_A#T zpe867bjXNi3wf2OgFf+Y?+26JcXu72f>2%)sc0F@p#u9BTOx2ijzX0nxkyM1(ZMnP0`(6=R@b8 zqB#1v(P$iB5YT*n*Un8L$~l5v8bb(m?!NO37J^COrsC<;aDg8val$uo^D#<6E1-{6 zbwhk_QOiUuSfi8t5AdJQvnxDE519#ycn`D_F$>5FzU-y{4@QtXrWB%E+t)YI0SS1; z+^rF;OgJ#f3Wy(Fk#~B)`(YLAD*+vh5qV8P+f0z519FHFGgH#58$brz;9L%Mh0v^1 zqN($QQ3V?{e@xEdgo;B=W6D8)F|u)baL#dWFwd|-lH^x3S2A=Tb5qp=$=nd_1Gp2B zIrtL+glwE>8sK4GExSM&BsLPb6C?Pag%M1&Jdoc<64gHNz$NH)7)zscw(26$JcrX? zwis%VX38;IcNA1g^=7QHWHs%Kbhn{=sd<#vM z##clGW4Ae)>!m5E2jIt(Sw4qqgj!D4N#$P!W+uONL{*S7^NL|2r{5zUB=XY@C*HA2 z8|31*d{tGhSqi+y9IZtS5*f|L|5Ke&D?$#7c^&crAIwFgD1+H)KQ*}&K=ePx(!4}y zWGXOssjm@b{Gu>WX)A^bR55CicI5%Jz7Bbsx*@E-!oLyhWUt;|rs&3s*vY{#Ow4|e zmrF_0&?4(6PKF47*<6fi^A;K;4nR(r8b>%V{)>%>4dPW-W#RWvb$tAm7}S~#3Nm;PW1oo;)TA11f>MWdFF4>8oq6G$^S3ALP+l@fH?OzD~cLhKT6Y|{6kT(zOiks zsOye-*P3mCru1%4^QJW)%RQ_sV=6|WdHNr9y~$o{VzZ&m$S%yBnsT~-RNjAmi3$ZJ z-u|Za$;~!@5L^?W=Yk4#yQ|x7CkQbedANq8vdkh76!J*VqoPmF(i_7xT%(sFK_gb<0d=za1Q1y1ax2N2 z|96J-aJ+gmSQz8?SqVhaO#w|+@M0mm!8(xqwx1c5BQzJ+scEPi2F~M6S4{nxJ2If2O+$XkrA?Y|x4?m-$ zJ3_PnKu~wmG7(#0Ll3|kK_>BLBGpav!G)F&~J7`5gTbl%H+!emBGgh6Fkc#d|to?TH~Ggd+JK@TQo4XZd?Q#7L`Xhv|?1 zUxg}>M<`GcMdHJ6z@rsEj@Ccm89UiE1&4AgR6f=Pw@U%26rt*1BFB zBBFrhu>o?PB#Hzc6ic?r>?#XVvyrtV)MVn%nJ3|PF{P8kVU@9;FMCnmqNH9CP1!H4Kc$Wj6emAC3yx}1?<}h zng5c@$FoqZKtiDI(Ci(4dW(Odjcqs8sf?pp(j4tRp}Gz`;!6%nwK4P5Ty@qCixAWC2@%vg`MI|%w7OCQM`r+bsWNg)`~nIp@@_D<5; z^QN*AXJdE=ij1boIRYx=V~#kxj`7JZjyH$pqTWY>>du71Iv`95YHv;ycth@0PMb-0 zHr4tJBVnxApdZjQP|S0KqCA|-y_|BY>FeSG{O-gI6 zjqKL{bsA-smcRw%A`Az5Ax3sxi|O!tiaEvN4Xu(F$OOA|e>%)Z0M_|N?A&gwQi`N& zEj8h952DnmLDt6eLR4S5P&BhgAMl4{eqGp817}EB-c&SVU6$42)@u41TBe*{O;9qr z2~=91*Ct#SRRUX2fAJ4iFV4+yQW7$zU;7Dtd=n|hvAHyCCji9_5A7+&j#=?BQugVz zArklZ`pxMgbJ1M2k`)*GJOZlPVQM#XBrZ(D^Y&^v^Oye~5*jAiORG%mJy8GTMT)!3 zGK1eLW<>QTkTyiX$6^&Q5oR%$w}8=B34QVh#+r}K=X7<=z8w2kjAmX8d)Qt8#pQMI zSeieHv*bG&pIu+(a#GQ{h0($O-XiOnE#}PAWXC%e=4e&CrC+}}+4&_+wTDEeROTXZ zkJ=DoeF^);CC$cg4Qd|A`9l=^umX^x=kw zh50j{K%SRreTv*MW0sW`0lPqEn@V6i#EmA=fdPb69ZTuYR}~#aZzx^xP_kYg7%kkr z4t$O;a;zWOr|ISPCP! zIv1yb`cL22Q>Haa`4ArLQPJ3w^kL%O_Fjn6w=MMr$;X_$@2!7Q-g4S^?bwD{Z4D>1 z3J5sDE{K=!`2HF9ndO3pf^@2v@90yXS3o165Go7tfqx|CTuRQZh$QGr zmV3NLc84|3C{e9QAVb}fO!|m*|GH5wDPU=LXcaSTZj-%R(8o8nM0Za@a7O_qVx%NeEW& zoOI7e;fw2w?E&J>{0-^#cT>sbT~=kua~i9PTc1B)iy^p|1AdZBJi~12XgM8W0E4!3 zDU~F?d9?|9*x4F<&0F@=t%O7Eg%*_5vGa4NuZB$1 zHB>1q`MZ>9R!PbB5md5{QFzL#1jR_%)xZB{phvI zJ-e<7tvE1^txvq)|C22YLIl>C8y$p0Rin`snhFMluLhlC#k%yXTd#yv7!}0s?1EZZ zJX`{SPrfPy*?+ZRBI+wL8!iE4ATSUCgvF|e4U5w!+IQB=r1UG9_N1KM<4rBUEX?ud zVQ^;9@WoeshI-F+jtv0V<&o$s&soW)piytDZP z?bH;9P?RE{SMk&W`07&=443l|bvw1J~OBM17vuPvC z>`;3xnP&a`Dm-ZQw_|imP)NFsuK(iggMq3c3^c%DiF5=j>>+gPR0~_K9-o_9!MnxK z@W)4!)l9`wF`c3ll>Ye6#TWyo(^EFNA?Jw@iJFkncZomd&|1(tM|D%Pl`= z5bvgS{%-+aPkkaFnayGa-t6mRNCH*;OQpQ99Rl=Ro__X0QH+@x>1sLSd}j5`*U&0h z&SGxG)k_PtS`=89JnLWE?^%agu+_A8y@o&v*D2Kbd9vYJ(cWp-k1be@7{y4idY-uH zUAl`TQV)BV+)R7YH%R4U)r3S9MG!6T??o(?YfdzZczx;k6ti?O`Odr^R@XY|*2e!f zSRf3Bn#yN-mz#oGz;1h5cbE%icb@vo;^pi`sYU#*bR9fr(9L{ggm|7pebm0{KT!mK z8jZiyu4J}U`yXG(iq4rsip#Jgd+jg0L^omYo$bNm?<%pZM{iu4QF1I2q90Jd-lRXB zC+W8+ZG{pk@?k~eDEH(2tVdq`Ro?6CX-LaAW<4*0-AEl1gI~r%%i&LaupqCp zydZiZh69JM+xFa{65WbBni>sq<_xNf$xD}kEOS0-L6E}sB7)$|0%D(hufMdAuX>)- zz~_ckO8Z2`S~0#<%JEx7%VVJ(Ty`ZAaxAI`$gPW$P2seXJbk|@pQhH$+hxaM zfL@4y_~=>SJ+(Aal%E+xaNYJb z(!%;@`-T*PkF2#-?wIRGvU?cpMSp+W4~5jBI6YIJsdh?tw@)a3ybAk$az8_81ionM zBA>6yXU?*+SN|!p$ZRUE4i+D)t*2{3wa8k7Ii3GQs&h)19l;bVTqcD+ACav#t|pB2 zWdF+ybaz@Kbayee z^H&aew?m%6gR8Nr0dga~KVoi64s)_2*bh&?3}k>f#m+^UHZ#c}-wWdy#iX)$dCLH` z7r}g8x%ZqVIkR=mm#SBX_yr>CS)pEp0FMg32^RusLu~S{gcQlR^NtTjR@{x>LJh5E z90?dIr`jpKnczCrShk=!OjBnws@o@& zV{g-vF+GW!D2I-}49vvMNS%t2is=oxx|_wFg*AvV)ejjSegb(7e|yw$@;1M3vh}m> z8rY8pmAUC|$HAXcbd`2au-sCeyj4FI*$6q!rcz!_DLjjDYJB#!XO)2rwt|>7^)(f` z#`8*vc<|n}bmvz`Btp8#&Xz|=LlGf=P$a9Dx4BSi=Ib)Pw*uPzZ%4}ZehoFZ$LT!p zVV&2PQoh=%kC4o+xEJ%<5t2%cYO6u|C<$Gh#+J41$Vt9gNn3P(+!BoX>ugVU4{lpd zHTQ1YNHkNjsosBLFW04M;SiB>6_cy5l9Wb4u`fX+OP`Yk+U;QWy7TDpS^IkU^S2Y< z9lPq{fpGna!rRlW^)wR~R9N25@FPa$s6?z{+2vJIvq#*o;MA+YiGJ9;EYT05f_7r>rk;{HRG};8;T|^S>_nTF7@LOVPl}u;5K4Di#udIJN^X!X2 z*h@3`k_SSYPPdVu%g1vy;rq(hGGAJ$re_pKSM%`8A^X|;o+hlc{6%<~F^!IZtFkQi zg)KTonRq9B62#qAB-=^suKq+$=_9x^7pLZ)5vVEGeW8xJuAj%PPQtPY3^*dZOZl#E zh%{m@#;%ZGUTqMEOw{ifb*rCsOtN7pmocclu#`%5$4WUR$diVID0_9b%TLlM{o3mj z5uH+9&p;7cH)Zs(NW7ZYUWsozkzbr`fzlNfi{OV?Kv^pChjv3U8)J9)+kD%PJY)87 zCM=HBqjJbtc6`FuFmzuSoevBfbb&GbSK(~ba#^V8OA59x4C%pb-TsQJb4QhRm!)*$ zMB|;#kp+9ASmK`%xMR$AZewMNUu3#t5Aez} zmaKN2kEFLWbI!^E(us4L0PMA^)dXM{Z^v_?z(&DML+|^*)i8_i1;d8%CEvk06^aLCiG*w zJvTx_If)ITKN~QU5i;_{><=xVO7mI6V3VHQ;_PclWz5dp6Ie=wQmx^CG;@!awZHbT z{=UTFinbz9Ok45;HQaE7t)6xTVK#sDmxaF-HeeQBE7yjNz2#O!F8&ngskrp=kxi}X z#@vLNB#USMy`8X0ub?Q%Z$Z~itZ_u})`DNx{c0?AFIb$bYR%fg)L62>Yh@+XG-Eew zLbxMuDJIEd_N?#CTi zO}cQ*U@W^_vkcxnT?KZf;jg3E+Wqc8VSI#}OtgMNBp<@$Yt%yXu5I7ZD; z=!=AS8!>bdk0qJ$grCTiy#X=K8w_!vRR7tp-ZTzk+~W%9lO0ChGZ)Io-qbxE%HoDo zGtmV(L@}|ur**GCsqH+*;_3_`oV=0!@v9}_OPk0Vu~!W2FK3xQXiaWj*LIXQn8!=r zuLxen&4f4sY}Sa2ge9qm7x7v4)18eI(soFF2&?@Ye{Z;aJx~DNOOQFVen&H>{`e>N z26Ndj zl5s;m?D(zx@R1dLm5Cd%Z)~o%6PJ8NAvn}uQYX((f<+^=QL=SLF?roqbENrm3c4ik zg!7}*UgUe}UMT(=rQF4s%H*X$hdEn=D*yI#zYNbh|(s4rSPFgoAoM8V1WX`6T zksnY^pfT$2VUg3Z$3GZ1VGSjpkpkrH#923jcLtm3c%Bdx?z>u+x>>(g#~nSNUwL5E zYgtZG=4T|-qfh|G89m_P<(IXZqCs8H6&d}WYwDS#ziTX5iKnj7XJ7cHhE8#}#1%(# zR^4t9cOWY;CFv2$0a1$9p+G=DVM&Wy1ggMXH<6Qa4pINv-yf)uvXXs-%?;7junolv zkx?&+&d{ZN=Km7!a;$WZ(1^cc2rsKe^Xi?6!PyUa5*{6ywr`lS`5rU}HZ)ePJ&{%> z?_zg;@c&t}^qGSE>N#E-^ z_&7eY%yjB?0@O!pGS?npN-+Uc4e8p3_y5) z*mVPoJ?i<@Rl2)__6r#hy~Ul13zirMIuu>Nz%4^W$EdS!#<%-+7?mSyy3Qqj0{0Cc% zT8&nE>NL3(BN3~HyC25GvN>Pg-rIkaHn`a63uKKZ9^CUfMF)M-gGDMi=V-)PR|;Fq zT;qa_wNsRM?3Y$Zu${Hlu*nM*qJx9#_F+5s>&z<|)7O5eM}scPu$8V}_K>b(39;xb z6O{P<7fE&HT+qz#EkkrQYCh3kc4}FRd#>mt8aC_=wMoB1nO&JlvId~pckDg zqnYb_JXroXgF{{Goyv~E`3L*IXZB#!H!x`(Z6Sur496s(E1|O;fGsM*tH=+G%ABR2 zcn_5dA@fRj=M}eB_(nbNnm1)w`5VXCRMGo2&90As)awET1`6}k9qtC!Thg(!FCE+V zoE&RRj+jnu|GOhZS>~y50TYDXGpSC1)~oH*N)+YB$~*^;sLI}h&8gIvn4Fk0|I(0w zeN?nizZDQ)80Hdl^~$yun&`b!fBu*s@YgY>ck>r>{+Ou%r`{Fw*ow&8sHveKP? z71hQm6I2$l5bsw5R3y8#zob1R@Js)ZPMO%7?!w0jGUPhxmsbTF2W6qsSgqK_2Yz&H zs4$7Ox2@rRn@{YdNT=THPS0>PN@dv7O9f-8#q>LfRL3UlxQ-1LPK?W!Nm!i+enpan za}P48hOu53>Q!z%#A_LEey((9WNu4#}7 z{p%q_hNzu(p?-H!{9smHWpd6)mYVp1U(-nZKt|%b*ZT7k6W|IOVuQ7hq8G-RLSy8B z8Q#9JNE&JVlhD!1lc|F8x|e}~a9F?w+zk0GC69TNO2}|O))VyPb;RM2S*O>MAUj zCORrm9TSf=eA28%y`%TFlt&bzQ%du3kL~bUAg;`@myi;**RHUTiVBk%^%WjUt(L)u zj<^$PQ&391O^hHxJb7J{Ni+ef6D`J}4Vm)kjvrAxm;2U&w`~Fp8k$2;E(4xLu&ajE zkr1jslO+w|&H(d8l}1wM7$@o<5L}kB3Jw&e##aOna%uEpuO4p~js-12FI#jZ6*F;*zn*rV?14k^CV7QLt9Ek?* zq>^=Mr@}#7_}uzdh~N7SrZ;P9FRazWqe@3iLdI=D#HYex@{Kg4-DQ^d_{3@_EM0_bR1U{C0dz@(@))e0lxr z!rCY?1LjV2+5TD47q8pLAupk!W@)+_ABXGWXa$B9(wQ#u%*9;X6{r|=CB4Ae>-8mA zMNg-dc=}cUbqOSUD?^EPXe`n|(Cqa4-5WW}rH?VzX7eN4B}}Bm_>}6&j%E*y6mZ3l zfB{9GFuzXg&h&3Odc$^EEcGX3jSi=BJIl8oB7ug!{jev2$><)yyS)+qT5hRK1Q^nw|X+4|A{Acp_s@R7eFS?sZDQMpJdt`oRB zw1xi2VRj71^IYsdN7O?PoQAhG!H;#lhYvJ5w9!JmIRBQOv9|y^PyNaqQl*Q(c~PnO z;aq#P)!BwP;j^^6=O@mWzJti3V+`HTEhhIm0agFRabuOJxr>m7e;yWzjbu{Y4$vtt z_9POrT>rhU-^&NNX1`##yB>-&Lxx&i6%4}shhM%Lr<}R`A1vpen!7vOd5ejer>|K%T%AMIN@*;uqnb*24c2wEt81 zRp_#Viw#Fa3u)ZQx>KOAr_mzxmFVDCO{s-f@Kv;%D;k}>e|JA7e(R2Is2zsz~rqJ*UO6t>?@0S*ZnsB$(#<9^6`YA}{elWW?ztOvzD z$mwWl4abDc>#%%Yv4OryIqP8kZT$s>fQCElPZ1MKG488~@L3Llr2lmz^9FoAd(*-t z5+8pxs`F`QK09m2YWq~F>?8rk+Jo@^QAV&tY~5jvKv{&9Ag%$&R)n|~ z2E*+$cd!P^cwpVy3n`ikzTLB;Iz(!s8|e-kA^z6lXgz;@X<79l8U4r-n<((xagYEX za~z1vQ_sJ9EIvio>1LWY@k&6@>)wpJCx?lNMPx8@Bo8UH@M);he^fP@+i+e-49ZkCY+?YTZZRqGJf~EWbLcym|IA%>ShjIM$YOWpF z#?mWLDd^45A>(O*xGxR{rNN@q1HWUUv6^DnNQbz%9BKgCmFv58JN}~()F4xOFFLA# zh;l8t!+P+9UTur%nOuj;z-wz|L>+o8V~Xwe8!QGU_U9j`OhESFE@lB0RGkj1$`y8n z&c8f+=m^R^9d4(0K%EyMh>)<1tMk5&rI&qTN0a}rnu1g|NP1}rZPZ^z)Gph7kDzl8 zK1vqCJG(Dcli)f8c`6!7p%@hB#75AYIl4+`TijBbX!v^_MhAuaM39r=Sa=5v zrDx)xtpeQ*eMGV9I#+vo>ym_*X6TScl+%Z3P1eS%AuR5PxV}(|Ah^TPVvUrcFwem0 zRbbK@HZZX&q}x3=CJN9Sv}j+dCMY{=+|%hEel5L& z@yBH0mmE+X) z@v~G}4J6nI%3ldgfnJJ>&9E_Bg>l^n45m$>&O^{hV7YsQiT-5xvoMzAPrOV{6fTh- zaeW`nK)=4vOUpAwL?LnhHgpP6ow1-^$No7^S=nMA8ZuBcD`{p1b_0v?ra8%8_W29H zVmKxgC7fr|f?u?xIXAi2?peanJ#9eAl;sFM*CsC_5Zw^<5~^+r=SYA$n~V|mtYTa# zWh)b@1GuZcU5pIq@;j@_PCXmb56LRVJjAgA&=4M^{OLUb7MkznqLm=s3C%cI{dACsMGxUvL|y*cKt1d zDGZUuKE`a=%Tr|ze#RVo0?o}QPiZLq#x|hQIIz-a4nTUJq5Bw^I(6$0$(0)_utNA@ zZa^JfNMQOzId4_-WPhQ&%@(&&l-51-UX69aQ>&IC$LwV_cB*`u+f%^I0Wl5)e>Qft z;ZRHjsH{@_%{5Ybs4LdQ^L@MBJN=6y=} zN|EP#Ex)Vi9-Cki>?JY6LuE?*KT z3mY*4H__vtRExo|Jz8E ztxJQBQCm|+L0J)uNM(5{=BovR{xMk^_0ql@Vx;3zb26;PqU6HMj6PS(zdvNlRnwa6Uv{bPWVXfdAhojWzg+zJT(Ka7 z7WYK=!z1FV9hy4^gB48)*0jyo&GDz72i3QH}?jEg4n6%lpUK zcdKXLwQ(N-9q4VU&YpM5`AB4cm1hymzLyiNhOLbbri8hzGrZL|vr0&7ATZB8(1lBe z(k+NIK3*zjDaI~F`aN7rpXF>-34^qdxD8Qi3QKZ%eHc=fXzDp z+X;3fW(AV+`bM!Z_64bXEgmpXPs1Qm*}QHPr=@Q)5iM05KrgMC6Jre;VLLPc&jQ#@X2H~2S!06{$zJZN!8vGt z(ElTh(ySqLze!oHMkDBvl`r#unT72Gmcwy*%h_kmrN5=@T?eB=F|)^S7*gxuW!VqD z3>F3Bdl=rVLZ9(y^+&{(`_=0r%`_>{T%l93QtnYaJfR$f56NSoqM=RXrsjDbcJjWa zm01sc;OUBGn!0+2y_!}=*{}49tF1F55OX698NUwx?@*@Ds&sNR64gw${J4hMfa?z?CFdbZ9Y@o&hX~49kW%F5D@Ik*y6`^UUF&rfb6y(b-#?6J8@Y5z zMFu$w>plV+VI8rz6_8)FFhaJIfj6 zr3bmcze%I0fgT>W=y%SzL}Sv_NXo-&iUOuy=0fW6^5#-X@$zKeXLJ~((P(T;HH68B zfi#rczEDKjLeHND1iI%w%Y$Hh?a8sU zK@MH??2RV8p$9JLx~V*qEPlSY?r4_}ajlW|{~O3xMHJqDB3Jjyh#-VCu*l`MiOYv{ zwvX0NdG`1rv!nAI8zWf(&M544U=i1Aso#w+LMVL1q|b|^)$Xrl!+rLfd9XnyAKSEC+ SlnXuPiJ_^k3#eBG#{M6y!Ig&q literal 0 HcmV?d00001 diff --git a/img/raven-logo.png b/img/raven-logo.png new file mode 100644 index 0000000000000000000000000000000000000000..e4c38b9a031176c9ae1b5a42f4ba89529b874893 GIT binary patch literal 10920 zcmXY1by(A17l(~*N$KuxX+cU5kd$UHx{)585>lc_qeBGAA>E@zfk{tt2+}bmm3oK2 z_n$p`?!D*S&&hkg_nhRX`dXw!3`AI1Sfr1&)s3*Q07{tuUkLCpzd*yMu~=9FK9AK^ zO+x1Pi`;UVoPIPT@<|dN3%zBu~WVpc@m{^P}w3-bdXnETi#Jz**?CO3x1VXyi*Bj zZ?Elcsd7T0Cz1wTap$AYkchv){^+sj{pdQNAaDUIWBWBBdpyO>lx{id))ialp$mG3 z=p~>V&y$g$u*@MNiY3INT#tY~K67h?p$v674?i=Pah?3?|7*jR0iOzXaU!v zyP{_*YKC~9xGfPek%b$*$J544W3br5NPR(UAvqsiw#A|)VYlXZF%+**|8cBnDG;EhB91E|#Y+U(;h16F;G5%J0KVgI5epIEOooG4 z&~`!sl60v%Ot!<_1GtC?BuR0ej@y4ad%%Ea4#*ka@s`1v?N0>OF0Nk_;3Ayhe(+84 zdd}Ye@NhYB<*3awAd><(^|8vavT6P=jCDPaFNc}GB-hz-P!1EGcDCg_Ij~j}_uD@m zk&o;K{dHI}QNh783-R>HFEI1KUG7h$R))qpVRWPh<^8n(h)~zbCcf;mK+}sU^pA6a@0y*F{tg`ZNF~(EB1?X4&jb}m>v~*pWGIRw~X|1_4S5Re@|tDg8AuxJWNhb$m7W`?&%SpuPQ_NGH^2P zz*D9ctHPfbGfzDky{(dbpxpN94r*WWmIZCkjx!HX6z?|YpF74=O*D|eUTZ-&El`xg zRd=-3sl}@orm~XB6=8Q0Q*dN$xg}rkn_q1q0l9dJP;Qz#pn?-rDKEj%MT$CJHpHV7 zuZZsWJ3%~1(8O0szZZbe<>z-v8<~rfEiBWf|gU6<3Hd&i(-We zyhqHIR}|kR0E%=Dr7%hh{l=Z?y9nXPfV$wKu*lDQ8ian{c@IIB_Q^{?Vrc8Fy&fVO zb*U!&=}yX5NyW@cOSfWu4TQ5gD;Yb+A1-k+o-N&J&+wu9e$e~?c)}PEgjI}F47f>e zf+28}ah8YMJ7{x~yD0Ow?>JK@=EU@C1)t|)!xhvuMniIcR?>Fm;dTP9fET=IYc%)W z?$q1-F89ALxU9|&9_EmDHvtC0|m$y@^bJ8RM%gYAdbQNr+1 z4;QT4K6$_)el2huFCAwH+vDB)TtAlQ0XG?lFGuOY0iEa00bS1m1Fl%PLQ0gnV-A61 zSNR(tCF{aeE2YS;G{1n_VG{vITe4P0WM|?*h2jKZH6_#KIJ}`A{LCZFZ;L9NwD#zz z4$A--MW}w(``QC<;W&9AE8lUB%-DWlh*u3QxlzwKDMh_5VYTuy!MIcxF;CmqxA|9{X?j@e4 zBZ^5ejRwZ+#0>IE$(}cQJsg!ju5GNEp@CpPE>h9#nJKH3nVVkzmFMPQ@2_uT`nHKO z?=CpnUlXH2#YQNpjNiZyTKQR9ZG3R^LcPv;Tx3tPrj@hfV>QF!YVFU!Gy4qIeA@Sq z+Bnzh-q&F-!vz?aK;-S^neE}P8+@y`SMULi&=j#mb2}iQh(FOlMkgHqMkQI6?i&a zYvsAug!G=nD$(H|zFUjq_9Wi4ZKRsZdKf9=vV1Qqk@M!_ZC0pRFRysTLn~9FIc)~Rwr4_L7>R-H-UXX|r8YCytL+RNchrBtm~Au;RFV$ zIXUfJ_VO#bm_5-xgxkC|xn&yP$ZB&v`+@U7LUfzKkaYp;Ek*)Wc29zCE7;~fp?^#V zvu8Q!h0LGbc_Vj0R_lj}jPERIir^YrbJ%dXiqce)bT(K1OZUOft(-nQf^;ky+##R} zk+*w^EXeL`;FcX^m0g@R)a}?$%;ytFBpcYcVF2_AZMEH}SqA1`ziIk3?1h9akERqS zw4bj+OLx^#e1QSK)V?%>`=>DJ63)Z7LUu)h#AP%I<@080PNvUJKe}3E(rNelZ6U_q zHJ8Q5LM#-=G?P%`xiC;qGp>P%YKj~;ir-|E1w*hd)P3mMN~6%LqW(~q^jWgGY~6AY=dYQGH&OHTuivIh9VNQTkjwG8 zMp9o8$QDWT`#*(O*VbHpmdvsbbyi2yN`*a`9~X{6-}~d_CgE5E0&#BG7=J>>F^?S@mQT&vp_rv1KwRTK}n= zQu&XQ#S66`_=)z9zF7O>dV7?Rjlp+Ug;5rZ)U7_MXGT^#^8zd{R(W)gPbVbFFWzz} zNy-15QV8j1E%h(5GV_I{{gIQRD6g+7 zayZ;=*uoC6*;SX^6Tq`)UpWoXc|QDbk-9K)Z^GlOgUrfOA!&ob=8>QC+2XpZLYVYO z!Tv)NxX`P53FjYcOdP>ej4#C%%7;@y~0D zkb$g<<$6N3U=;+1HAex{3wrr5LC6hhXX>QLSmlugaVv3w9Du<#7hm-fy!b&?h0AUO zI%O{L>Po`f-R84{n!Gfp8JpG5A>D05sS`6~^!Rd!0K{466R82(11nz1eiEU{fb0`)B*NiS`41uvIKCWVVSXA3=yY-e`R*5zx(B+}nfi z?NimC#=tREaV2i<{xWmFrBI$)3t{uwYx8sO*Di!jf|r7hY#aDt?VEMkq41(~Tb~XY zSJq3^&`cN3Vr@&}=yPdr7r~;Kg1(S)C@+ah7kk{ONjzd1p5X3;ppK{)zecPM`KOv& zMlN7o_G2XdUcf)Ad$;E#raP25RiZ{^2V&t$ODPamHI40JjyPeRIh!Z9C`Z_#c7 z2(v%yO#~NAf&!7G#OJp7g(RL!?-=*iimAmrBW#yy@x`z?#@obk&$oc_`y2!X^g0Rg$$j z!Nc!aw#2}3;Em))VD9U*XwxInrobYRU)D!NXwDZ43xMFP^#N%m`wy%3i|a)i5q;&H zt=o%9O^v!2)NxbwC-`#wpl;wr{CrU>YbGJT8;unCa9Yu@!o^es-=a8uNbl;INFUw_ z#PAD=SqHj#HcyJ!RUyuxy2W1kL4MI|AKU(=_@5tSd%RrtJXzcekx#y!YNt7-xBw6{ zI2#^;{F|@HhMD8K>YGD+)Ddb@U~B`Zk07vTu&{RIVQ@;tEmc@p`lyD>r2j&~WrdQ2 zx-8F!f{^ZIQ9zwMa@@Ujn`MzT&?~w2kHT!*PD(voxjTsk_THYjY1|gZGka9eM2r0j zJwjIwh#Jh8-eD&r&eaVJ&+q*B5O-`Yl(Ni1lp^K9{mM@~x#9kJ=#TyTHmYDv(P?go zE%w0#!&<7%Ag%dUs}Eba$x1f4xBu04a(shb;!kg=NB)urCOQMKT!#|X7W zO&2`C+dmu-ERj0jW18)z&xi3%4OK8XRz>Z#G0w;K&?8RxOZnTgz(w`0&mgtGwWVNY zfae=j*NSP)&{+Ei;$A6~1I8xINI9M60@`6km*;D58MXrP>oLD)ER3 zcybEdDaRP5>buXfKc)WSV}7s#Mf^UlID6NRRZ#CTf9)l;&1`~$vUf*YPX9d8SkEV+ zZZ-^4H6!=Iio)9LTfb*Qf8@PW!XsqauhTf|Ds7_6TqEaZscm)=oJ4%z_3zgt zIf)ykox~d^&?8Y2{E*JPdF#HyEFXLsvovyLy=Tf~-;8guWR-Q(prQvm^40}qwuYJ1 zbKx)gLBZu`3@OY^iYr2%u|0eu=MJT^)<;U!ETS_fwDnBOjy2-S8sYFng{345f}|6A z;zIJ#5qS4OdO@fRbE}U-v~#fcWJ++v9sej}r#pRCF7*uZnD_LSJgq(JYsbEO_!s_P56WDF;>w~q~KT}s1XG5seu{b;C`$CLFZ zLs5t9iciD~*L}k_L&K{VjIH*ygbS%U4;W4ZQ~C+CZ)54CGWsU-ABUQ-LU3pBUn}qU zhN%p?n;BCn`H2kZZCSJ)ZUL$eEey)zi#N*b0vc=5>c`oiTf+;vA>dOp4GJZyto?ZH zjB93DdA&fJO2fLXP}WwR-Cf~tSZ}FFg^cYbrOSK`d|SKhLd>5@-$&NG+YPJOygX0g zoalP)8AMk$d@-)j6)WlvnS3QYN`0QL6IbbC-xKqcbqes*(TpTeu z(LDPu5`RxKl1T9PDGZ`&Yz_8XbBs_p7^X_P^1FwY^8SX~y&ks}I(0mcJN61&}dyU;Xy zft+Y6pP$=y;~fL0H%n8uFq1w;QkZw3Ja#(>w=dyTt&hQxoEPU&jbEmKr+ic-D6OR@$=qU8Yk3=TKz_7LWi1Og$I@xm}sP+$Q zsa>X5F(PVod)pyPQ8jPV-))vY5&|i@8^I_CS&Q1ER0Cnc8xG&uT$Iut%jhmsLotkrl^=rS^ zyq^XmR<=a@mkM*IsF%=*$+(aGY+gbYIU>e!mcOKKy|1p>v#qJ4tHvv=n99tR^nEt$ zDNUS58KNdZMUYRG(!5py`0umZ^llxrFVem=_YCnV?k9X%=k zG}wMSFI~f=0f`;F0SyJ)UCFRt1JTL&mwwr{0)6+6naPEp*qpp}=-U)Qh3znKjIG8q z*F!hT8fm$N35tHXTCKU-x@+K$2K-pU0CnR%;E|g976JF{GKqh56Q9;Lq6ZOV(QC~= z+03TSli@(Tzoyw|cS}q7whi5su`hZ1h?dKOF2RkC!eByIT`HOti}jbr$xRv@3Dfd^Lo>bgOO?_WO<_d4<_s4q0f0#Hib|} zvDS$9hC%suQ$&Al|N1d-N?R5nm6lUL=bW}W9GKp)^ z;li~a@8gac8*6{!XAsTr^=duJKh`;UI$It^AT#hZVN7c2h!#Zv(vyOT5e_R9Pc9-> z&xXMJ$OreCqlgE$ynipnpWtCGJWT##=ule?a(Gqlxt`V!MS9hKE2IKUq?StH11ql5 zPG#G$y~iK%i%mgk?u*QylB}!ahdwG98UC~~D<9}PlY02|vJ8~rJ7LfHA$^Q{kLqRr z%4meUog}2~)^b%QEe5^UwRo%CpPttdd{A2u#7K*Z7_>1>TBs1|6uVk zB^2EVB^{$F!cxh(=f+{%TO?{2<(WfSOf6Q*HH|-DhgKZ*j*;06k?#wAid$pK1Az zWM+RADNFzjb$&g9dUL1>@aH={ZLG0%rWh;fQ>8xOCA^8$TBKh4i~6IkBN=u!=w&EJ z@2~i_v|YPE7iLJ`hDr4-gF>5A)OH$9dW#=SQ$`J9f?i}L&KRW|J<$V$%++j%^yF*F zVZ4#$zc*yjU1)uqaSkV)v>*qc(+@-IOi+=R<2)%vIk7#ha3hY=*xU12X)PsS^LS8e z*W+nfkPmN)c)<%it*~%?zow~HaX<_3I6#h=Jfta$b;wGo?I;J9pP1LH94PhJbF3Vx zF}4o}p(>CCmG4AYYuZ{18X8!(b5px@g39}s@6OLU-jehKwShl!q?PBVWntwsyNPl| z8;$iUX>Fa}>1A^%I(4W^9F$f2_>MEnEa{6!VMIo`WB#Fub0D$<`ND7+_6U-ygn|0q z)42ARxRKI)d)Y1%#wiIEhdQ~CO0FDQ9O%v@YD1Y;aypEKaavZ?#jVPS9I9k5fApX_ z&{AXU)Q_L`Q$TMJJV#nNh^l&tVa(M}reW9x^R93}JXY?mU5(D%ZWU%m>thZT%ixvW z-jzTc*#mtcB`WVt5g`>7GIK_Ax(gOFk_*CAgSjp)Ez#E+a0__Bv>7K&H(vA523G~V zpnX@^dgDE!v`w0?1<)Ybjn^}*M?DZsHMxt4P3m~b=s8R>_$fOt8$F^#q2tx-*XWSY z>R*pO@it%I3Wi%Pg};(x4Jo;5#k*zHT^S=B*L5Lx9u&ePoQbh5>BYSM1C#!z{Sj~G zIhMVnNdFl4U2}$%3`}^wb_v)f%20Wt|D|dx*`Isqe5VCeBx!(!a3~fJ1}aP)i=Dq% z_n_SFH9xq0hm@bA?6SaZQGTjXn*C*tL<@J)#-=n(Zl^VD_`ScP++)A}mn%FUYN1L! zvrdAFn|1JnP&%c3^OQvIN#WapzV974B`L$kQTD&H9OvS$!h*`FsR# zZhwNQGiaYkWn9+hDVorKZ}$9gbYgPlX|w8biyfJy7w0*K$~_ZPYGs5u{|F6@Y*cxYh@x%6tJv`f;Rh%`_3Qrw~j7*DjB-N9rj7`mtl8vm7+RGK9vp?0m ziE0#~`Hu)}_ND~EBLT6dW|{&ClzRwVdm4iViF(NOu33h!R3H*4gzQ;HQ}J*DLQd#F z!L0H(gQ3@5X(zduXR_`q;x;bSBW7l~I$6&#;=XxLV*8uAh7T*#`0x_&?XBDds!vEh zGlOWrhC6PK71r7i@hN5ZT-!zjQ#Rs)ua`6Yz$V1t0jrH`D~t4u31C9>-Go^Cr?PaSq~{(0nzHS{xrtyw^Q3`0h(R z68;FI2sLPV3<-Akca0Qx0_W}!USCPw*{!`~`YaZs+3ss;P!np%Hsw#KUU@t3Oh3KN zZm&uDZ2a)GrEesyS>8eet)XfoMkVYMQ*4RAh{~j5IrXd}_k?dnNW~Jtn6Ts>V;Nq7 zXfmtGL*Vkqzwm*_j$vxYlQtjfZ<1y%UK2ybX!s^DWiXD06AlGsqZy*SsMMMLLR==B zGV>Pcjp-ZV(Y$LpPy;fQfb|lBj!1aO7A}5qo&w(yvZW}Lp2ZIuWQ6mpn=bYnC}KLKWnZmv z`a{WJ@a8sb)Mp2$T9doe@U469993}!yE$=~T{Wt$DfQ=$P;xXZBrL4bN_PMk#UHcv zr@G*n?G9FKv$8Jvx4y_5EJO;nI4r-(-J~Aa>AaVJe@6BEL*W}A0)5Nh8 z=#L1;J(Ii@Du`28I_X9S?ge4aO{K9C1Y$?j@jdv_a}!Z5Nz76 z?d`VwG`;A^hP^8$M#0Xr_suQ_o@7}VJsE~f81JUF#pi_=zX*MB8pxHTiL+2MutwOt zkXYt54j8%@s{f|EBL`U9Mh8t75sW( zu6usZ+~s=;d;k@kFr<=cykL;swij7#~xHk!x3{ZSnOya`ur1AUj^Gp|lS+Z}d zHm`Vn6*kDC$Y(91!qAF#xH;~qzb*K?XS8N<)8DekB`|i2btdP1IoIl$_{Y*qW6n6` z(kwUTm|>RVdYENz!K^ne*IcKPRM%_&WmB9i)QSY;Q;pl@^8c&{)v`px|(@aQqcY!(1=3_52>D_KdxVc-4%RjPu9FA+Q_cjhw zO^3ErzjctE|C4)p=30%FD}jIPF8li_h2D_S1CO^fI&V`;%97es)c)ciZlC@j?J(@y zY=PJI`Zt3N??uM}?rlO^!F&`7--v4L@VW^X=b~2=SN*Hg%c$<+b!?lz<8gNqjU+3u zla3(szu9OP5ttRi9_I1EDJ2Y6>`Wij$F{pKYGzD|Mt zVSMxJY18^ZK{@?x;nHuacjpP>RUh}dTDAF~c82CTBj+Pu>{*^}I>UUS^(29Xkk=Il zAF^yT{d)ae*uMYda0#$DDbKo89pyTq5YlLc+j|*&Qgb)L@OdahtxJ@bQE&2H1?F8U zdiKzyI#xqWhM0r2OHB3zD0nqyE%aF`b>lKhiHe}fP#qVZ0owmHw)~biS_RDi$&|_a zoK`577n3BP4jj~2rrP&&to^F5_#$q;Q`wPTD&|G~`?1iN(M@phy|OoxXe`EP`aJQuKCpk-mc*zue01*zi&binke&=(19@8w_L4n@jrxJQ zfppNI<)Z9!F_(=6a|U5A9+1XjIV#eOf~;_A4zAcQkX(F=w?gM2t;PzTkY<>s6vc=_ zt$7k{q3~qDJ@C~Bd*wkRxw#L4wOj<483H1!FN;OmuKn#bH$;or;lK;lon5h=)R~FK z&lew@^BhG&mp!iUde5GIYcHtK&b6gaq>;_ghIRQ`l>=oR_?e8Jv}P#S!fl$)U&Mp_cs1!2MOA~_Cypob7U92tyW|>fufeb3VB~17ju1Xma#I!g$KPBou$;3Z;Oc+< zH|CgU-`A1wRr3HBOb6f_1CTeqEgUL$4))qCIY)(R0aO4(!_9WKmy*P(54tK`|8+1~ zt3zh{h+YR*tk&PcNhDbNpUAuLE^aE->%&ez75;{>T-bq}+1lls{$42sii*YnkDnmbmNKZ9EKci%E9$-JHPeHd4rxFBe-{o5hfP29gCOSr5;4?s+ zLoA0#$VcpKF}q4Tuwijq?;1qzehQs_=#40?G=xYuj5KZEg*anIy+D zOj}nNv-F*|fh-BW{vZ#R3PADkwEF=#3;hs93F0s`Je`Z4Yh1j8mU0r%Yd!jC_Z~_Z zbkA_?N%@Or24Yy_0*9xwLC-^&QspmN%(BdL+NFXdBN^B4!!qb`^YU2)EL4dVBz%CG ei{}%&W%W;!1i$SR7sYg-Vm;Q-SFcqA#r_X{o~YRX literal 0 HcmV?d00001 From 647ecae96dc72ad196c9b8ef5ad415a61a1f17bd Mon Sep 17 00:00:00 2001 From: Eduard Kerkhoven Date: Sun, 1 Nov 2026 16:20:28 +0100 Subject: [PATCH 2/3] docs: use the Jaccard range of the validation study --- README.md | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/README.md b/README.md index c29b685..cb0f866 100644 --- a/README.md +++ b/README.md @@ -31,7 +31,7 @@ The in-memory model is a [`cobra.Model`](https://cobrapy.readthedocs.io). raven- ## Status -raven-toolbox 3.x tracks the RAVEN 3 pre-releases and is a pre-release itself ([releases](https://github.com/CompMetEng/raven-toolbox/releases), [changelog](CHANGELOG.md)). On Human-GEM, the five cell-line models of Hart2015 extracted with ftINIT have a Jaccard index of 0.975 to 0.980 against MATLAB RAVEN ([Human-GEM validation](https://raven-docs.readthedocs.io/en/latest/parameter-tuning/studies/humangem-validation.html)). +raven-toolbox 3.x tracks the RAVEN 3 pre-releases and is a pre-release itself ([releases](https://github.com/CompMetEng/raven-toolbox/releases), [changelog](CHANGELOG.md)). On Human-GEM, the five cell-line models of Hart2015 extracted with ftINIT have a Jaccard index against MATLAB RAVEN of 0.964 to 0.982 without tasks and 0.981 to 0.988 with tasks ([Human-GEM validation](https://raven-docs.readthedocs.io/en/latest/parameter-tuning/studies/humangem-validation.html)). Three functions of RAVEN are not implemented: From 9fc0088c6493e343f30fac7a8881c090ff919b89 Mon Sep 17 00:00:00 2001 From: Eduard Kerkhoven Date: Sun, 1 Nov 2026 16:20:43 +0100 Subject: [PATCH 3/3] docs: README wording --- README.md | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/README.md b/README.md index cb0f866..e2214bd 100644 --- a/README.md +++ b/README.md @@ -27,7 +27,7 @@ Reconstruction, Analysis and Visualisation of Metabolic Networks in Python: the ## Design -The in-memory model is a [`cobra.Model`](https://cobrapy.readthedocs.io). raven-toolbox has no RAVEN struct and no adapter for one. Fields that cobra does not model, such as `rxnMiriams`, `metDeltaG` and `rxnConfidenceScores`, are kept in the `annotation` and `notes` dictionaries of cobra, so that raven-toolbox stays interoperable with the COBRA ecosystem. +The in-memory model is a [`cobra.Model`](https://cobrapy.readthedocs.io). raven-toolbox has no RAVEN struct and no adapter for one. Fields that cobra does not model, such as `rxnMiriams`, `metDeltaG` and `rxnConfidenceScores`, are kept in the `annotation` and `notes` dictionaries of cobra. This keeps raven-toolbox interoperable with the COBRA ecosystem. ## Status