# The Evolution of Brain-Derived Neurotrophic Factor (BDNF) Reveals Insight into Brain Disorder Vulnerability

Alexander G Lucaci^1*, Michael J Notaras^{2*^}, Sergei L Kosakovsky Pond¹, Maarten van den Buuse^{3,4,5^}

<font size=2.5>
¹ Institute for Genomics and Evolutionary Medicine, Science & Education Research Center, Temple University, Philadelphia, Pennsylvania, USA. <br>
² Center for Neurogenetics, Brain & Mind Research Institute, Weill Medical College, Cornell University, New York, New York, USA. <br>
³ School of Psychology and Public Health, La Trobe University, Melbourne, VIC, Australia. <br>
⁴ College of Public Health, Medical and Veterinary Sciences, James Cook University, Townsville, QLD, Australia. <br>
⁵ Department of Pharmacology, University of Melbourne, Melbourne, VIC, Australia.

*Michael J Notaras & Alexander G Lucaci Contributed equally and share first authorship.

^ Shared corresponding authorship between Michael J Notaras & Maarten ven den Buuse, email [e-mail] or [e-mail].
</font>

`Updated on 3.9.2021`

## Abstract

A comparative molecular sequence evaluation of 319 species with genetic orthologs for Brain Derived Neurotrophin (BDNF) were examined across the taxonomic range for jawed vertebrates (Gnathostomata).

Brain-Derived Neurotrophic Factor (BDNF) is an essential mediator of brain assembly, development, and maturation which has been implicated in a variety of brain disorders such as neurodevelopmental disorders (e.g. autism), neuropsychiatric disorders (e.g. depression, PTSD, schizophrenia), and neurodegenerative disorders (e.g. Parkinson’s). Loss of BDNF during early development is embryonic lethal, and depletion of BDNF during adolescence and/or adulthood can result in disease-related neuropathology across model organisms. To better understand the role of BDNF in disease, we seek to provide context to BDNF’s role within the brain by elucidating the molecular-genetic evolution of BDNF across species. First, we conduct phylogenetic reconstruction of the BDNF gene/protein? across a diverse selection of vertebrate species (n > 300 species, and n taxonomic orders) to examine differences across clades and taxonomic orders. Next, we examine sequence alignment to determine genetic distances, unique genomic attributes across species (e.g. GC content, branch lengths etc.) to determine how BDNF has diverged across and between species. Evolutionary change was next considered via two independent computational models, validating novel sites within the BDNF prodomain and mature peptide coding region for synonymous and nonsynonymous changes. Lastly, we applied branch-site statistical models to determine changes across lineages within the phylogenetic tree, revealing that at least 24 species exhibited differences in their selection pressure across their respective lineages. Of interest, humans were not one of the consensus sequences to indicate significant selection pressure within their lineage, which is indicative of tight evolutionary conservation. In sum, our computational genomic analysis of the BDNF gene estimates the age of the BDNF gene as likely emerging xxx million years ago, and that BDNF has exhibited different evolutionary selection pressures across different species within organismal lineages.

**in sum, these data evidence an evolution of BDNF towards unique species-specific regulatory capacities, but not/more so than an entirely new ontogenesis of BDNF function.

## Introduction
Brain-Derived Neurotrophic Factor (BDNF) has become amongst the most ubiquitously studied molecules in modern neuroscience ref. BDNF is a neurotrophin that binds with high affinity to its cognate tyrosine kinase receptor, TrkB ref, to elicit rapid induction of synaptic plasticity ref and spine remodelling ref. Additionally, BDNF has been implicated in a variety of brain disorders, including depression ref, PTSD ref, schizophrenia ref, Parkinson’s ref, and autism spectrum disorders ref amongst many more. BDNF has correspondingly been the primary target, or an ancillary factor, of many therapeutics currently in development ref. Yet, nascent research has provided the humbling reminder that much remains to be discovered about BDNF. In recent years, new BDNF ligands have been discovered ref, new receptor interactions unveiled ref, and mechanisms of function unlocked ref. This is a timely reminder that while BDNF has remained a seminal molecule of interest across the broader neuroscience literature, much remains to be discovered about its origins, function, and disease relevance.

A Primer of the Molecular Biology of BDNF and its Functional Topology
BDNF is encoded by a single gene of the same namesake (BDNF) ref, and is regulated in humans by an antisense-gene (BDNF-AS) ref that can form RNA-duplexes to attenuate translation ref. The BDNF gene comprises 11 exons and can produce at least 17 detectable transcripts ref. Different transcripts are induced as a differential response to activity and/or cellular states ref, allowing the BDNF gene to be highly responsive to environmental stimuli and potential selection pressures. However, despite producing multiple transcripts, all transcripts ultimately yield a singular preproBDNF protein that (prior to intracellular processing, cleavage, and transport) can be partitioned into three domains ref; namely, a signal peptide, a prodomain, and the mature domain. The signal domain is only 18 amino acid residues long, exhibits ambiguously defined functionality, and the majority of BDNFs functional outputs reflect sequence specificity to the prodomain and mature domain. The BDNF prodomain encodes binding sites for intracellular transport of both BDNF mRNA and BDNF protein ref, and contains numerous post-translational modification sites ref, is the resident location of the most widely studied Single Nucleotide Polymorphism in neuroscience (Val66Met, or rs6265) ref, and the Furin consensus sequence for cleavage to its mature form ref. The mature domain of BDNF is composed of, almost exclusively, the NGF domain and is responsible for the canonical trophic actions associated with BDNF (e.g. long-term potentiation, rapid-acting antidepressant effects etc.). Following intracellular handling, processing, and transport, the preproBDNF isoform is cleaved to yield the mature BDNF peptide (which only contains the mature NGF domain). For many years the prodomain was thought to be cleaved following transport and thus destined for degradation.

For a comprehensive, detailed, analysis of the various intricacies of the BDNF gene, protein, and its regulation, more information is provided in ref.

*The Conservation of BDNF & Neurotrophins*

One of the interesting curiosities surrounding BDNF is its relationship to other neurotrophic (NT) growth factors, comprising NGF, NT-3, NT-4. Thus, neurotrophins retain some intercalated functionality. Each share some commonalities in structure (pre-, pro-, and mature-domains) ref, post-translational modification potential (e.g. glycosylation ref), catalytic processing and trafficking, and composition ref. Specifically, neurotrophins share approximately 50% sequence homology ref, and a comparison of domains and motifs exemplifies that each comprises a prototypic NGF-domain as the principal component of the mature pro-growth peptide for each factor ref. While each neurotrophin elicits functionality via binding to cognate receptors, NGF, NT-3, NT-4, and BDNF each exhibit cross-affinity amongst neurotrophin receptors ref. Not surprisingly then, there is some redundancy in the pro-trophic effects of neurotrophins, yet each still maintains nuanced functionality which remains specific to each factor during central nervous system development ref and homeostatic regulation of circuits and behavior in adulthood ref. Differences in the evolution and temporal dynamics of regulatory sequences, which target gene-products to specific destinations within cell-compartments ref or to processing routes ref which alter secretory dynamics and/or bioavailability ref, likely contribute to both similarities and differences between neurotrophins.

Purpose of Study
One unexplored idea in the neurotrophic factor literature is that evolution may have shaped the sequences, structure, and function of BDNF. Indeed, there is evidence to support this idea. First, BDNF shares remarkable similarity with its neurotrophic family members, yet in spite of this they are unable to compensate for BDNF should the gene be deleted entirely (i.e. it is embryonic lethal) ref. Second, BDNF appears to be highly conserved across mammalian model systems, namely rodents and humans ref, despite being clades remaining rather evolutionarily distant (in relative terms). Third, genetic variation which alters the BDNF peptide has been demonstrated to be coupled with altered function as well as a suite of disease states, indicating that loss of conservation may be inherently maladaptive ref. Fourth, BDNF has been shown to be a locus of substantial stress sensitivity ref, raising the possibility that this locus may be similarly coupled to environmental selection pressures. However, the genomic origins of BDNF have remained undefined, and no substantial computational analysis has been performed to examine the phylogenetic evolutionary history of BDNF, it’s selection pressure sensitivity across lineages, nor quantitative metrics of evolutionary change across species.

Purpose of Study
Here we use computational methods to explore the evolutionary genomics of neurotrophic factors, beginning with BDNF. By reconstructing phylogenetic trees of BDNF in jawed vertebrates (Gnathostomata), we utilize sequence alignments of hundreds of species to determine genetic distances and unique genomic attributes of BDNF. Likewise, utilizing multiple models of evolutionary change, we identify novel sites within the BDNF prodomain and mature peptide coding region that are susceptible to synonymous and nonsynonymous change. Lastly, we identify that the evolutionary selection pressures on the BDNF gene are species dependent and specific within lineages. Together, these computational evolutionary analyses provide important context as to the origins and sensitivity of genetic changes within the BDNF gene, which may be important for providing context to genetic risk factors linked to disease in humans.

This work additionally provides broad comparative genetic insights into the evolutionary history of the BDNF gene family. Our set of novel, identified polymorphisms in key functional regions of BDNF may provide significant areas of interest for designing molecular therapeutic approaches.

## Results

**Color key**
-> Blue: Mammalia
-> Orange: Sarcopterygii (lobe-finned fishes)
-> Purple: Teleostei (ray-finned fishe)



### Figure 1. Inferred Maximum Likelihood Phylogenetic Tree for 319 genetic sequence for BDNF. Visualized via iTOL. Cladeogram and Tree with branch lengths

Maximum likelihood phylogenetic trees inferred with IQ-TREE (Minh et al. 2013) from the BDNF coding regions of 319 species encompassing the gnatostoma (jawed vertebrates) clade. The scale bar represents the expected number of nucleotide substitutions per site. Highlighted (see color bar) are several large taxnomic groupings () maximizimg phylogenetic diversity. Phylogenetic reconstruction of the BDNF gene, across diverse selection of species (>300 species across n taxonomic orders). Maximum likelihood phylogeny of BDNF sequences from this analysis

Origin & Phylogenetic Reconstruction Across Species

TTL Calculation

Branch length distribution

Origin & Phylogenetic Reconstruction Across Species

### Evidence for recombination
We did find evidence for recombination in a subset of sequences. The most pronounced event occured in the following sequences (Event 1)

| Common name | Species | Accession | Start | End |
| -------- | -------- | -------- | -------- | ------- |
| Sunda flying lemur |Galeopterus variegatus | XM_008584165.1 | 380 | 1733 |
| Rhesus macaque | Macaca mulatta | XM_015114598.2 | 380 | a. |
| Crab-eating macaque | Macaca fascicularis | XM_005578346.2 | 380 | a. |
| Drill | Mandrillus leucophaeus | XM_011964715.1 | 380 | a. |
| Sooty mangabey | Cercocebus atys | XM_012031697.1 | 380 | a. |
| Sumatran orangutan | Pongo abelii | XM_002821931.2 | 380 | a. |
| Silvery gibbon | Hylobates moloch | XM_032166370.1 | 380 | a. |
| Olive baboon | Papio anubis | XM_017948537.3 | 380 | a. |
| Gelada | Theropithecus gelada | XM_025358305.1 | 380 | a. |
| François' langur | Trachypithecus francoisi | XM_033205475.1 | 380 | a. |
| Golden snub-nosed monkey | Rhinopithecus roxellana | XM_010374529.2 | 380 | a. |
| Angola colobus | Colobus angolensis palliatus | XM_011936307.1 | 380 | a. |
| Black snub-nosed monkey | Rhinopithecus bieti | XM_017858805.1 | 380 | a. |
| Green monkey | Chlorocebus sabaeus | XM_008003703.2 | Text | a. |
| Tufted capuchin | Sapajus apella | XM_032283691.1 | Text | a. |
| Panamanian white-faced capuchin | Cebus imitator | XM_017538722.1 | Text | a. |
| Nancy Ma's night monkey | Aotus nancymaae | XM_012452238.1 | Text | a. |
| Gray mouse lemur Gray mouse lemur | Microcebus murinus | XM_020286534.1 | Text | a. |
| Common marmoset | Callithrix jacchus | XM_002755152.4 | Text | a. |
| Canada lynx | Lynx canadensis | XM_030333208.1 | Text | a. |
| Northern greater galago | Otolemur garnettii | XM_003781198.1 | Text | a. |

Other events were noted in the following, however, because they did not have at least 3 sequences available, they are left out of subsequent selection analyses (The analyses will not run on <3 sequences.)

| Common name | Species | Accession | Start | End |
| -------- | -------- | -------- | -------- | ------- |
| White-footed mouse | Peromyscus leucopus | XM_037204438.1 | 405 | 1791 |
| Southern multimammate mouse | Mastomys coucha | XM_031371313.1 | 1 | 380 |
| Chilean tinamou | Nothoprocta perdicaria | XM_026041160.1 | 671 | 1320 |
| White-throated tinamou | Tinamus guttatus | XM_010226640.1 | 671 | 1320 |
| Ugandan red colobus | Piliocolobus tephrosceles | XM_023230347.2 | 384 | 617 |

Recombination free alignments were generated. This results in the main alignment now having 319 sequences (recombination free). And a second alignment consisting of 21 species with sequences from Event 1. We will analyse the main alignment and separately the Event 1 alignment in subsequent selection analyses.

The full alignment consists of 1791 nucleotide positions.

They will be refered to as the recombination free main alignment (Main)

and the recombinants from the first recombination event (Event 1)

### Figure 3. (BUSTEDS, gene level)

| Alignment\Parameter | N | Sites | Global Omega | LRT | p-value |
| -------- | -------- | -------- |-------- | -------- |-------- |
| Main | 319 | 597 | 0.149 | 9.975 |0.0034 |
| Event 1 | 21 | 597 |0.122 | 0.6250 | 0.366 |



### Figure 4. (FEL, site level analysis) Inference of dN/dS rate estimates
Main

Main, Positive Sites
| Site | alpha | beta | alpha=beta | LRT | p-value | omega |
|----:|---------:|--------:|-------------:|--------:|----------:|----------:|
| 34 | 0 | 1.73056 | 1.14873 | 4.27353 | 0.0387103 | inf |
| 134 | 0.606648 | 2.36183 | 1.88689 | 3.95668 | 0.0466857 | 3.89325 |
| 187 | 0 | 1.28625 | 1.15896 | 5.77227 | 0.016281 | inf |
Main, Negative Sites, only showing first five sites for brevity, 232 total
| Site | alpha | beta | alpha=beta | LRT | p-value | omega |
|----:|---------:|----------:|-------------:|---------:|------------:|----------:|
| 41 | 1.81046 | 0.0972841 | 0.683871 | 7.15214 | 0.00748746 | 0.0537344 |\n
| 43 | 2.42935 | 0.372006 | 0.899683 | 4.75852 | 0.0291537 | 0.15313 |\n
| 45 | 74.341 | 1.11598 | 2.21325 | 11.5957 | 0.000661052 | 0.0150116 |\n
| 128 | 4.22923 | 1.02237 | 1.71537 | 7.5426 | 0.0060257 | 0.24174 |\n
| 180 | 3.65396 | 0.496403 | 0.973044 | 5.80839 | 0.0159499 | 0.135853 |

Event 1

Event 1, no positive sites returned

Event 1, negative sites returned, only showing the first five for brevity. 24 were returned in total. **Note:** besides site 129, they all have a beta of 0.

| Site | alpha | beta | alpha=beta | LRT | p-value | omega |
|----:|---------:|---------:|-------------:|--------:|-----------:|----------:|
| 129 | 12.5707 | 0.945439 | 2.92175 | 4.36859 | 0.0366073 | 0.0752097 |\n
| 240 | 4.9428 | 0 | 0.618239 | 3.98112 | 0.0460129 | 0 |\n
| 280 | 4.91853 | 0 | 0.546246 | 4.22191 | 0.0399053 | 0 |\n
| 314 | 6.07471 | 0 | 1.79518 | 6.77373 | 0.00925094 | 0 |\n
| 315 | 12.8583 | 0 | 3.568 | 9.38812 | 0.00218396 | 0 |

### Figure 5. (MEME, site level analysis) Inference of dN/dS rate estimates
We find little evidence of adaptive evolution with FEL, so we turn to a more sensitive analysis.

Main
| # | Site | alpha | beta- | p- | beta+ | p+ | LRT | p-value | num_branches | MEME_LogL | FEL_LogL |
| ---:| ----:| ---------:| ---------:| --------:| --------:| ---------:| -------:| -----------:| ------------:| ---------:| --------:|
| 1 | 26 | 0.161755 | 0.123455 | 0.983392 | 765.359 | 0.0166084 | 9.36072 | 0.00406614 | 0 | -25.8914 | -22.2119 |
| 2 | 28 | 0 | 0 | 0.760612 | 6.38989 | 0.239388 | 5.16213 | 0.0346796 | 1 | -30.4859 | -28.6386 |
| 3 | 34 | 0.0731002 | 0.0391922 | 0.760523 | 9.46854 | 0.239477 | 7.61631 | 0.00987126 | 1 | -34.2615 | -32.5901 |
| 4 | 39 | 4.25978 | 0 | 0.90773 | 219.758 | 0.0922696 | 5.96379 | 0.0229696 | 0 | -47.6588 | -42.6318 |
| 5 | 49 | 0.0704958 | 0.0380175 | 0.839385 | 22.0613 | 0.160615 | 5.35347 | 0.0314264 | 0 | -35.873 | -34.3692 |
| 7 | 91 | 0.628558 | 0.303951 | 0.932209 | 205.003 | 0.0677907 | 29.2336 | 1.83785e-07 | 2 | -122.008 | -107.524 |
| 8 | 92 | 0.709668 | 0.623679 | 0.764322 | 14.3349 | 0.235678 | 6.44114 | 0.0179871 | 0 | -122.139 | -119.75 |
| 9 | 97 | 0.302341 | 0.035981 | 0.832711 | 64.6297 | 0.167289 | 5.61364 | 0.0274918 | 0 | -92.0292 | -88.0246 |
| 10 | 98 | 0.90578 | 0.0632336 | 0.862902 | 28.2386 | 0.137098 | 8.55769 | 0.00611371 | 0 | -92.7062 | -86.7974 |
| 6 | 81 | 0.5513 | 0 | 0.908234 | 17.7701 | 0.0917657 | 4.98037 | 0.0380857 | 2 | -79.1138 | -68.863 |
| 11 | 100 | 0.503264 | 0.279114 | 0.942441 | 689.446 | 0.0575594 | 12.9678 | 0.000655263 | 0 | -90.6863 | -82.9598 |
| 12 | 102 | 0.0558658 | 0.0360955 | 0.833407 | 22.7748 | 0.166593 | 5.56265 | 0.0282217 | 1 | -76.0316 | -73.356 |
| 13 | 125 | 0.552809 | 0.0270208 | 0.795705 | 11.7654 | 0.204295 | 5.84005 | 0.0244748 | 2 | -94.2607 | -89.4859 |
| 14 | 126 | 0.0572224 | 0.0406115 | 0.795361 | 55.3515 | 0.204639 | 18.8638 | 3.35915e-05 | 2 | -125.727 | -117.127 |
| 15 | 127 | 1.58782 | 1.58782 | 0.926015 | 1851.01 | 0.0739853 | 5.52066 | 0.0288373 | 0 | -119.75 | -117.183 |
| 16 | 128 | 3.76605 | 0.327006 | 0.951837 | 484.612 | 0.0481629 | 8.34118 | 0.00682517 | 0 | -94.9832 | -88.4232 |
| 17 | 129 | 0.336927 | 0.336927 | 0.885992 | 749.501 | 0.114008 | 19.4909 | 2.45052e-05 | 0 | -122.696 | -113.181 |
| 18 | 132 | 2.35604 | 0 | 0.81369 | 18.5801 | 0.18631 | 5.34335 | 0.0315905 | 0 | -99.2216 | -94.8981 |
| 19 | 134 | 0.255333 | 0.255333 | 0.881651 | 69.3742 | 0.118349 | 18.3158 | 4.42546e-05 | 2 | -104.473 | -97.2943 |
| 20 | 135 | 1.29445 | 0.599529 | 0.880116 | 49.118 | 0.119884 | 5.35249 | 0.0314422 | 1 | -103.08 | -100.381 |
| 21 | 136 | 1.27383 | 0.310808 | 0.883208 | 57.5919 | 0.116792 | 8.78645 | 0.00544271 | 1 | -84.5894 | -77.4551 |
| 22 | 142 | 1.42739 | 0.734231 | 0.945094 | 88.1313 | 0.0549065 | 5.55357 | 0.0283536 | 1 | -92.374 | -89.1927 |
| 23 | 143 | 0.719601 | 0.0748341 | 0.881351 | 41.0882 | 0.118649 | 10.9905 | 0.0017804 | 2 | -83.7961 | -77.1508 |
| 24 | 145 | 1.71172 | 0 | 0.750942 | 11.5018 | 0.249058 | 5.12458 | 0.0353571 | 1 | -86.8917 | -83.5688 |
| 25 | 150 | 1.65778 | 0 | 0.916812 | 30.3827 | 0.0831883 | 6.328 | 0.0190594 | 1 | -38.9506 | -33.3234 |
| 26 | 187 | 0 | 0 | 0.822842 | 15.0537 | 0.177158 | 22.9086 | 4.39778e-06 | 22 | -242.642 | -220.361 |
| 27 | 199 | 0 | 0 | 0.921627 | 76.6077 | 0.0783731 | 5.03515 | 0.0370248 | 0 | -22.6223 | -21.4494 |
| 28 | 214 | 0 | 0 | 0.897003 | 0.616839 | 0.102997 | 5.26364 | 0.0329133 | 2 | -24.5805 | -23.4518 |
| 29 | 354 | 0.38257 | 0 | 0.989733 | 370.876 | 0.0102667 | 10.7409 | 0.00202016 | 0 | -29.0425 | -22.9415 |
| 30 | 355 | 0.449345 | 0 | 0.982187 | 20.48 | 0.0178128 | 5.3946 | 0.0307684 | 1 | -25.3093 | -21.4219 |

Event 1
| | alpha | beta- | p- | beta+ | p+ | LRT | p-value | num_branches | MEME_LogL | FEL_LogL |
|----:|--------:|--------:|---------:|--------:|---------:|-------:|----------:|---------------:|------------:|-----------:|
| 130 | 0 | 0 | 0.853302 | 63.457 | 0.146698 | 6.5093 | 0.0173706 | 3 | -21.9007 | -19.6036 |

### A closer look at the NGF domain

Corresponds to site 468 to 575 in our alignment
FEL/MEME positive sites failed to fall within this area.

FEL Negative sites in this area for Main
| # | Site | alpha | beta | alpha=beta | LRT | p-value | omega |
|---:|-------:|----------:|----------:|-------------:|----------:|------------:|-----------:|
| 1 | 468 | 1.4683 | 0.0370125 | 0.416971 | 60.1498 | 8.77076e-15 | 0.0252077 |
| 2 | 469 | 1.35294 | 0 | 0.362645 | 64.2769 | 1.11022e-15 | 0 |
| 3 | 470 | 3.05122 | 0 | 0.793955 | 147.019 | 0 | 0 |
| 4 | 471 | 0.672082 | 0 | 0.189614 | 35.5138 | 2.5325e-09 | 0 |
| 5 | 472 | 0.58717 | 0 | 0.0900441 | 22.2017 | 2.45457e-06 | 0 |
| 6 | 473 | 0.671018 | 0 | 0.254039 | 35.7535 | 2.23924e-09 | 0 |
| 7 | 474 | 4.59521 | 0 | 0.44023 | 142.146 | 0 | 0 |
| 8 | 475 | 2.23286 | 0 | 0.542227 | 99.464 | 0 | 0 |
| 9 | 476 | 1.35294 | 0 | 0.255424 | 65.1007 | 6.66134e-16 | 0 |
| 10 | 477 | 0.0927779 | 0 | 0.0135239 | 3.90107 | 0.0482553 | 0 |
| 11 | 478 | 0.526846 | 0.0149764 | 0.109466 | 20.8965 | 4.84776e-06 | 0.0284265 |
| 12 | 479 | 1.27444 | 0.0692717 | 0.365324 | 44.6253 | 2.3859e-11 | 0.0543547 |
| 13 | 480 | 4.56722 | 0 | 0.419939 | 137.196 | 0 | 0 |
| 14 | 481 | 0.875004 | 0.0555052 | 0.173521 | 18.3203 | 1.8671e-05 | 0.0634342 |
| 15 | 483 | 1.35294 | 0 | 0.31259 | 55.4814 | 9.4369e-14 | 0 |
| 16 | 484 | 2.18314 | 0 | 0.567256 | 108.606 | 0 | 0 |
| 17 | 485 | 1.73111 | 0 | 0.458437 | 89.4829 | 0 | 0 |
| 18 | 486 | 1.69111 | 0.0936624 | 0.521881 | 57.8406 | 2.84217e-14 | 0.0553851 |
| 19 | 487 | 1.25506 | 0.118636 | 0.299427 | 26.0262 | 3.3682e-07 | 0.0945264 |
| 20 | 488 | 0.497179 | 0 | 0.111794 | 23.6792 | 1.13803e-06 | 0 |
| 21 | 489 | 0.294264 | 0 | 0.0449957 | 11.161 | 0.00083533 | 0 |
| 22 | 490 | 1.10526 | 0 | 0.329165 | 60.9799 | 5.77316e-15 | 0 |
| 23 | 491 | 0.943664 | 0.017771 | 0.293253 | 45.662 | 1.40525e-11 | 0.0188319 |
| 24 | 492 | 0.322277 | 0.018927 | 0.0946596 | 10.0843 | 0.00149542 | 0.0587289 |
| 25 | 493 | 0.293079 | 0 | 0.0406705 | 11.7529 | 0.000607469 | 0 |
| 26 | 495 | 2.91693 | 0.0195794 | 0.748449 | 121.19 | 0 | 0.00671235 |
| 27 | 496 | 1.35294 | 0 | 0.329536 | 62.7782 | 2.33147e-15 | 0 |
| 28 | 497 | 1.55792 | 0.0564591 | 0.371707 | 49.8169 | 1.68787e-12 | 0.0362401 |
| 29 | 498 | 2.24526 | 0.0545426 | 0.573518 | 85.6901 | 0 | 0.0242923 |
| 30 | 499 | 0.739912 | 0 | 0.225322 | 39.1267 | 3.97173e-10 | 0 |
| 31 | 500 | 2.0544 | 0.0547258 | 0.56193 | 79.8728 | 0 | 0.0266383 |
| 32 | 501 | 0.524583 | 0 | 0.160222 | 27.9857 | 1.22218e-07 | 0 |
| 33 | 502 | 1.62804 | 0.0402516 | 0.509713 | 65.2949 | 6.66134e-16 | 0.0247239 |
| 34 | 503 | 1.85714 | 0 | 0.338476 | 75.6144 | 0 | 0 |
| 35 | 504 | 1.96296 | 0 | 0.323357 | 75.58 | 0 | 0 |
| 36 | 505 | 0.610115 | 0.0192199 | 0.199841 | 26.0488 | 3.32892e-07 | 0.031502 |
| 37 | 506 | 1.63436 | 0.0164177 | 0.460828 | 81.4746 | 0 | 0.0100453 |
| 38 | 507 | 1.35294 | 0 | 0.355366 | 62.2247 | 3.10862e-15 | 0 |
| 39 | 508 | 1.25487 | 0.190954 | 0.473546 | 27.0294 | 2.00385e-07 | 0.152171 |
| 40 | 509 | 0.800274 | 0.0173994 | 0.17975 | 30.4306 | 3.46031e-08 | 0.0217418 |
| 41 | 510 | 0.287363 | 0 | 0.0883685 | 16.4054 | 5.11381e-05 | 0 |
| 42 | 511 | 1.35294 | 0 | 0.276538 | 54.284 | 1.73528e-13 | 0 |
| 43 | 512 | 0.215667 | 0 | 0.0924799 | 11.7477 | 0.000609182 | 0 |
| 44 | 513 | 0.300803 | 0 | 0.0451394 | 11.2875 | 0.000780299 | 0 |
| 45 | 514 | 0.621874 | 0 | 0.137197 | 26.8366 | 2.21407e-07 | 0 |
| 46 | 515 | 0.675354 | 0 | 0.11744 | 23.9776 | 9.74638e-07 | 0 |
| 47 | 516 | 0.148805 | 0 | 0.0253714 | 7.02891 | 0.00802041 | 0 |
| 48 | 517 | 2.51828 | 0 | 0.417373 | 83.8247 | 0 | 0 |
| 49 | 519 | 0.0927779 | 0 | 0.0249738 | 4.61552 | 0.0316839 | 0 |
| 50 | 520 | 0.510956 | 0.0360756 | 0.12979 | 13.8524 | 0.000197747 | 0.0706041 |
| 51 | 521 | 0.605505 | 0 | 0.0813138 | 23.6852 | 1.13453e-06 | 0 |
| 52 | 522 | 0.616957 | 0.0485951 | 0.136485 | 14.5627 | 0.000135571 | 0.0787659 |
| 53 | 523 | 0.873852 | 0.0692561 | 0.286504 | 29.4951 | 5.60575e-08 | 0.0792538 |
| 54 | 524 | 2.931 | 0.170733 | 0.308716 | 29.1187 | 6.80782e-08 | 0.0582508 |
| 55 | 525 | 1.01154 | 0 | 0.295894 | 55.0743 | 1.16018e-13 | 0 |
| 56 | 526 | 0.572788 | 0.0403327 | 0.133137 | 12.4295 | 0.000422595 | 0.0704148 |
| 57 | 527 | 2.25059 | 0.0360041 | 0.594403 | 95.2355 | 0 | 0.0159976 |
| 58 | 528 | 0.63564 | 0.0357556 | 0.133595 | 15.9245 | 6.59198e-05 | 0.0562513 |
| 59 | 529 | 1.7249 | 0.19835 | 0.391125 | 24.4319 | 7.69849e-07 | 0.114992 |
| 60 | 530 | 0.956466 | 0.0555505 | 0.315311 | 34.9183 | 3.43833e-09 | 0.0580788 |
| 61 | 531 | 0.50745 | 0 | 0.0677883 | 19.8247 | 8.48776e-06 | 0 |
| 62 | 532 | 1.75862 | 0 | 0.434189 | 87.8365 | 0 | 0 |
| 63 | 533 | 0.357816 | 0 | 0.110289 | 19.8699 | 8.28962e-06 | 0 |
| 64 | 534 | 0.529313 | 0 | 0.076082 | 23.2027 | 1.45794e-06 | 0 |
| 65 | 536 | 1.35294 | 0 | 0.19871 | 49.0664 | 2.47435e-12 | 0 |
| 66 | 537 | 0.432881 | 0.0528124 | 0.146711 | 11.0491 | 0.000887308 | 0.122002 |
| 67 | 538 | 2.75357 | 0.0473198 | 0.389632 | 81.6587 | 0 | 0.0171849 |
| 68 | 539 | 0.97289 | 0.0403751 | 0.111611 | 13.0581 | 0.000301976 | 0.0415002 |
| 69 | 540 | 1.58536 | 0.0159977 | 0.223004 | 50.0213 | 1.52089e-12 | 0.0100909 |
| 70 | 541 | 1.35294 | 0 | 0.358598 | 66.6795 | 3.33067e-16 | 0 |
| 71 | 542 | 1.00173 | 0 | 0.1767 | 37.5923 | 8.71874e-10 | 0 |
| 72 | 543 | 0.702343 | 0 | 0.0949889 | 27.6001 | 1.49174e-07 | 0 |
| 73 | 545 | 0.0927779 | 0 | 0.0356412 | 6.33878 | 0.0118126 | 0 |
| 74 | 546 | 0.392851 | 0.036294 | 0.140314 | 12.8993 | 0.000328711 | 0.0923861 |
| 75 | 547 | 0.796293 | 0 | 0.1292 | 28.4758 | 9.48784e-08 | 0 |
| 76 | 548 | 0.893394 | 0 | 0.267918 | 47.2714 | 6.18083e-12 | 0 |
| 77 | 549 | 6.25489 | 0.0407657 | 0.676369 | 127.165 | 0 | 0.00651741 |
| 78 | 550 | 0.529313 | 0 | 0.151529 | 27.1849 | 1.84901e-07 | 0 |
| 79 | 551 | 0.495767 | 0 | 0.201805 | 26.7279 | 2.34213e-07 | 0 |
| 80 | 552 | 1.35294 | 0 | 0.334734 | 64.5055 | 9.99201e-16 | 0 |
| 81 | 553 | 1.31576 | 0.0563887 | 0.415288 | 49.4864 | 1.99751e-12 | 0.0428565 |
| 82 | 554 | 0.374837 | 0 | 0.113475 | 21.275 | 3.97878e-06 | 0 |
| 83 | 557 | 1.89932 | 0.122216 | 0.323808 | 38.6662 | 5.02851e-10 | 0.0643474 |
| 84 | 558 | 0.369514 | 0.0354962 | 0.111525 | 9.69867 | 0.00184401 | 0.0960619 |
| 85 | 559 | 0.833024 | 0.054893 | 0.180777 | 18.4223 | 1.76976e-05 | 0.065896 |
| 86 | 560 | 1.66282 | 0.0362856 | 0.389245 | 60.9917 | 5.77316e-15 | 0.0218217 |
| 87 | 561 | 0.898587 | 0.0923569 | 0.324281 | 26.3561 | 2.83927e-07 | 0.10278 |
| 88 | 562 | 0.527623 | 0.0374953 | 0.182383 | 18.9461 | 1.34465e-05 | 0.0710647 |
| 89 | 564 | 1.35294 | 0 | 0.449822 | 72.1418 | 0 | 0 |
| 90 | 565 | 0.945446 | 0 | 0.1649 | 44.51 | 2.53064e-11 | 0 |
| 91 | 566 | 0.623239 | 0 | 0.0891851 | 27.0657 | 1.96657e-07 | 0 |
| 92 | 567 | 0.89956 | 0 | 0.201681 | 44.1408 | 3.05588e-11 | 0 |
| 93 | 570 | 0.169178 | 0 | 0.0590615 | 10.5002 | 0.0011936 | 0 |
| 94 | 571 | 1.35294 | 0 | 0.338504 | 61.9138 | 3.55271e-15 | 0 |
| 95 | 572 | 0.229253 | 0 | 0.0497017 | 12.2097 | 0.000475409 | 0 |
| 96 | 573 | 0.434904 | 0 | 0.151531 | 25.1573 | 5.28405e-07 | 0 |
| 97 | 574 | 0.19493 | 0 | 0.0381511 | 9.75054 | 0.0017927 | 0 |
| 98 | 575 | 0.21287 | 0 | 0.0720341 | 12.9564 | 0.000318832 | 0 |

FEL Negative sites in this area for Event 1
| # | Site | alpha | beta | alpha=beta | LRT | p-value | omega |
|---:|-------:|---------:|-------:|-------------:|---------:|------------:|--------:|
| 1 | 470 | 3.8868 | 0 | 1.2806 | 4.39173 | 0.0361137 | 0 |
| 2 | 474 | 4.9428 | 0 | 0.546279 | 4.22377 | 0.0398614 | 0 |
| 3 | 480 | 20.2048 | 0 | 1.71581 | 13.6398 | 0.000221442 | 0 |
| 4 | 483 | 5.16321 | 0 | 1.53914 | 5.81029 | 0.0159326 | 0 |
| 5 | 484 | 11.7536 | 0 | 2.51621 | 10.0125 | 0.00155482 | 0 |
| 6 | 485 | 5.24063 | 0 | 1.27113 | 5.13438 | 0.0234565 | 0 |
| 7 | 487 | 9.85713 | 0 | 1.27435 | 7.83053 | 0.00513711 | 0 |
| 8 | 500 | 4.31365 | 0 | 1.19137 | 4.64326 | 0.0311757 | 0 |
| 9 | 507 | 3.68557 | 0 | 1.15132 | 4.21369 | 0.040099 | 0 |
| 10 | 508 | 5.20006 | 0 | 1.60807 | 4.1805 | 0.0408915 | 0 |
| 11 | 527 | 3.14997 | 0 | 1.10545 | 3.99338 | 0.0456794 | 0 |
| 12 | 528 | 3.9162 | 0 | 0.696912 | 3.98954 | 0.0457837 | 0 |
| 13 | 542 | 8.65167 | 0 | 2.01006 | 6.37964 | 0.0115437 | 0 |
| 14 | 547 | 8.05973 | 0 | 1.76675 | 5.73587 | 0.0166218 | 0 |
| 15 | 549 | 17.7234 | 0 | 1.67296 | 7.23815 | 0.00713705 | 0 |
| 16 | 553 | 4.33289 | 0 | 1.5105 | 4.10905 | 0.0426543 | 0 |

## Methods
### Data Retrieval
Query NCBI via https://www.ncbi.nlm.nih.gov/kis/ortholog/627/?scope=7776
336 full gene transcripts and protein sequences were downloaded.

Table of species included in this analysis: https://github.com/aglucaci/EvolutionOfNeurotrophins/blob/2021/analysis/BDNF_lineages.csv

### Data Cleaning
We use protein sequence and full gene transcripts to derive coding sequences (CDS) (via a custom script). However, this process was met with errors in 17 "PREDICTED" protein sequences which had invalid characters and these sequences were subsequently exempt from analysis. This process removes low-quality protein sequences from analysis which may inflate rates of nonsynonymous change. Original number of sequences 336, errors found in 17 sequences so 319 species will be considered.

### Alignment
Codon-aware alignment performed via MACSEv2 [ref] with default parameters, see pipeline for invokation.

### Recombination Detection
Manually tested via RDPv5.5 [ref] with modified settings as follows:

* Recombination events where 'accepted' in cases more than 2 methods agreed.
* Slightly modified default parameters
* Sequences are linear
* List events detected by >2 methods
* Alignment was save as a distributed alignment (with recombinant regions separated).
* This single fasta was separated into two files the first a recombination free file with 319 sequences (with recombinanted regions separated out) and a second recombinbation files with 21 sequences with these removed sequences from Event 1.

### Phylogenetic Tree Construction
In all cases, maximum likelihood phylogenetic tree was constructed via IQ-TREE [ref], see pipeline for invokation.

### Phylogenetic Tree Visualization
https://itol.embl.de/ [ref]

Link to shared tree: https://itol.embl.de/tree/173687017153241614097877

### Similar matrix heatmap
Constructed in R via RStudio Cloud
Project link: https://rstudio.cloud/project/2230943

### Software availability
All software used is available via a dedicated github repository at: https://github.com/aglucaci/EvolutionOfNeurotrophins

https://colab.research.google.com/drive/1J87aXP3iv2_0iTFGaG8ZSdp-i0yUZ-Oj?authuser=1#scrollTo=bPt1pv4qj1vu

Processing BUSTEDS jsons via https://colab.research.google.com/drive/1BXoup4TEMFT8AwFG_VdZIOF6RadM41bj

### Mapping Human sites to the alignment columns

https://colab.research.google.com/drive/1SanrGHfpAknXLjKP9-Ix1xjUMRunWJzE#scrollTo=rv_UqFIJoy0R

### NGF Annotation
Human BDNF
https://www.ncbi.nlm.nih.gov/protein/767966315
region is 215..322 in the reference
in our alignment this correspondes to a region between site 468 and 575.

### Tables generated
Are all available at: https://github.com/aglucaci/EvolutionOfNeurotrophins/tree/master/analysis/CSVs

## Discussion

This indicates that within species, unique evolutionary pressures and changes within the BDNF gene have evolved across time. Of interest, humans were not one of the consensus sequences to indicate significant selection pressure within their lineage.

Of note, BDNF elicits tight regulation and specific functionality that can be separated from these other neurotrophins, yet these growth factors remain closely related in their structure and sequence.

*Limitations*

This analysis focused on gnathostomata (jawed vertebrates), not invertebrates (clade exclusion, other BDNF gene or BDNF-like analogue gene may be in other claves, but not explored). Evolutionary temporality is also important in this context.
Species exhibiting high recombination were not included because ______________

*Future Directions*

We hypothesize that the similarities between neurotrophins reflects positive evolutionary selection for motifs and domains which support common functionality between neurotrophic factors between sites and lineages. While we note significant isotropy in mature peptide sequences for these factors, anisotropic pressures likely influenced the prodomain sequences of neurotrophins leading to alterations in processing, trafficking, regulation, and secretion. As such, we also predict differences in the evolutionary fate of neurotrophins which compartmentalize functionality, namely due to alterations within the prodomain of NGF, NT-3, NT-4, and BDNF.

## References
1. https://journals.sagepub.com/doi/10.1177/1073858418810142

## Supplementary Material

## To discuss:
* Species specific regulatory regions/capacity of this gene.
* Overwhelming negative selection in NGF domain.

## To do,
1. Annotate with reference 1.
2. Sitting on results from aBSREL, FMM, PRIME, BGM
3. Also SLAC
4. Disease relevant sites from humans in NGF?

### Figure 2. Genetic Distances among all sequences (Pairwise sequence comparision in the alignment). Heatmap of distances is constructed as a similary matrix. Tamura-Nei 93 (TN93) pairwise genetic distances were calculated in the codon space.

Protein Space

Codon Space

TN93 - Codon Space

Differences in sequence alignment (pre vs pro-domain, statistical differences in genetic distance, genomic attributes in GC content across species, branch lengths etc.). How significantly diverged across BDNF sequences between species. (protein-space)